(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Climate in everyday life"

1 




UNIVERSITY 
OF FLORIDA 
LIBRARIES 




Digitized by the Internet Archive 

in 2012 with funding from 

LYRASIS Members and Sloan Foundation 



http://archive.org/details/climateineverydaOObroo 



■w 



CLIMATE IN EVERYDAY LIFE 



By the same author 

THE EVOLUTION OF CLIMATE 

CLIMATE 

BRITISH FLOODS AND DROUGHTS 

THE WEATHER 

CLIMATE THROUGH THE AGES 



CLIMATE 

IN EVERYDAY LIFE 



By 
C. E. P. BROOKS 



LJC. 




PHILOSOPHICAL LIBRARY 
NEW YORK 



551. 5 
Bg'li 

ABCH£ 

tectuke 

SOOK ROOM 

Published, 1951, by the Philosophical Library, Inc. 

15 East 40th Street, New fork 16, N.T. 

All rights reserved 



PRINTED IN GREAT BRITAIN BY J. AND J. GRAY, EDINBURGH 



PREFACE 

THIS book sprang from a short paper which I contributed 
to the Royal Meteorological Society in 1946, entitled 
"Climate and the Deterioration of Materials." Requests 
for reprints far exceeded the supply, and it seemed that a small 
work on this and other applications of climatology to industry 
might serve a useful purpose. So I began a book on "Industrial 
Climatology," which, by a natural process of assimilation grew 
into the present book. 

The literature of the subject is so immense and so widely 
scattered — much of it is published in trade and technical 
Journals — that I cannot hope to have seen more than a small 
part, though I have been greatly helped by those workers in 
technical research who have kindly sent me reprints of their 
papers, and by the Librarian of the Meteorological Office, who 
has called my attention to other new work. Moreover, no one 
man can grasp all the technical complications of modern life, 
and I have no doubt made some howlers for which I ask for- 
giveness from the specialists. 

Acknowledgments are due to the Air Ministry for permission 
to reproduce the anemogram in Fig. 12, to the Royal Meteoro- 
logical Society for Figs. 14 and 17, to the National Smoke 
Abatement Society for Figs. 20, 21 and 22, and to Dr. J. M. 
Stagg for permission to use the material of table 9 in advance 
of publication. Some of the other illustrations have been 
reconstructed from figures or diagrams published by other 
workers; the source of these is acknowledged in the text. I 
am indebted to Miss E. H. Geake and Dr. J. Glasspoole 
for criticism and advice on climatological matters, to 
Commander Hennessy for help with the section on sea ice, 
to Miss N. Carruthers for reading the whole of the text and 
playing the role of candid friend, and to Miss M. E. Robinson 
for making fair (I should say, fine) copies of the maps and 
diagrams. My wife has, as usual, helped in all stages of the 
book. 

C. E. P. BROOKS. 
South Ferring, 1950. 



CONTENTS 



PAGE 



Preface ......... 5 

Introduction . . . . . . . . .11 



PART I 

Living with the Climate 

CHAPTER 

I. The Economics of Climate . . . . .17 

II. The Siting and Design of Houses and Factories in 

Relation to Climate ...... 56 

III. "Temperate" Climates of the Old World . . . 95 

> IV. Temperate Climates of America . . . .115 

V. Sub-tropical Climates with Summer Rainfall . . 130 

VI. Tropical Climates . . . . . . .151 

VII. Desert, Mountain and Polar Climates . . 159 

PART II 

Climate as an Enemy 

VIII. Climate and the Deterioration of Materials . . 171 

IX. Atmospheric Pollution . . . . . .188 

X. Climatic Accidents . . . . . .201 



PART III 

The Control of Climate 

XI. Heating, Air Conditioning, Lighting, Clothing . . 233 
XII. Altering the Weather . . . . . -258 

appendices 

I. Climatic Tables ....... 277 

II. Measures of Humidity ...... 293 

III. Conversion Factors ...... 294 

References ......... 295 

Index .......... 303 



ILLUSTRATIONS 



FIG. 



1. Types of climate . . . . . . 19 

2. Average temperature distribution, January . . .24 

3. Average temperature distribution, July . . -25 

4. Mean wet-bulb temperature, January .... 28 

5. Mean wet-bulb temperature, July .... 29 

6. Operating time and power yield of a windmill at different 

mean wind-speeds . . . . . . -37 

7. Level of a rain-storage tank, London, 1938 39 

8. Number of months with snow-cover on low ground . 45 

9. Diagram for height and azimuth of the sun ... 65 

10. Effect of cloud and skyline on illumination ... 78 

1 1 . Illustration of turbulence ...... 85 

12. Specimen of anemogram (reproduced by courtesy of Air 

Ministry) ........ 87 

13. Mean annual rainfall of Australia, inches , . .113 

14. Radiation to and from a container with thin walls . . 172 

15. Temperature inside a container heated on top surface . 175 

16. Probable maximum temperature of a surface exposed to 

the sun . . . . . . . . .177 

17. Absorption of water- vapour by leather . . . .182 

18. Rate of growth of mould at different humidities . -183 

19. Index of deterioration . . . . . .186 

20. Section across London, West-East, to show how pollution 

affects winter sunshine . . . . . .191 

21. Sketch map to show distribution of total solid deposits 

(metric tons per sq. km.) per annum, over open country 

and to windward of main centres of pollution . 1 93 

22. Composite map of pollution expressed as a percentage of 

that at the centre of town . . . . . 1 94 

23. Distribution of pollution in Glasgow . . . 195 

24. Suspended impurity at Kew Observatory . . .197 

25. Maximum rainfall in one hour (inches) expected once 

in two years ........ 202 

26. Maximum rainfall in one day (inches) expected once in 

two years ........ 206 

27. Number of lightning flashes to ground per square mile per 

year, estimated . . . . . . .217 

28. Distribution of tornadoes (vertical shading) and tropical 

cyclones, hurricanes or typhoons (horizontal shading). 219 



10 ILLUSTRATIONS 

FIG. PAGE 

29. Frequency distribution, hourly temperatures, Kew 

Observatory, April . . . . . . -237 

30. Ogive or cumulative curve of temperature . . . 238 

31. Construction of cumulative curve . . . .239 

32. New York cumulative hourly temperatures, computed and 

observed ........ 241 

33. Comfort and danger zones ...... 246 

34. Wind speed up- and down- wind from a wind-break . 271 



INTRODUCTION 

PRACTICALLY every action of human life is directly 
affected by climate. The food we eat, the clothes we wear, 
the house we dwell in, the work we do, are all dominated 
by the climate in which we have the good or bad fortune to 
live. Not only is our life governed by weather and climate, but 
so also are the energy with which we live it, and the good or 
bad health which we enjoy. The efficiency of any group of 
people, such as a school or college, a body of professional men, 
or the personnel of a Government department, factory or office 
is closely bound up with the climatic conditions in which they 
work and live. A stimulating and health-giving climate is 
actually worth" a very large sum in solid cash, to say nothing of 
happiness, to the country fortunate enough to possess it. Not 
all countries are so fortunate; many parts of the world are hot 
and enervating, others have long monotonously cold winters. 
But these conditions should be regarded as a challenge rather 
than as a discouragement, for much can be done by suitable 
housing, air conditioning, clothing, sanitation and in other 
ways to minimise the ill effects. This has been shown in such 
places as the Panama Canal Zone, once notoriously unhealthy, 
where wise organisation has brought life and health up to a 
level little inferior to that of the centres of civilisation. 

A word is required here about the distinction between 
"weather" and "climate." "Weather" is the aggregate of 
atmospheric conditions at any one time in any one place; it 
changes from day to day, while \ climate," which is the whole 
long-period assembly of weather conditions, goes on all the 
time. Our day-to-day activities are more or less affected by 
the day-to-day weather, but the general pattern of our lives is 
governed rather by the climate. The city dweller who goes off 
to work every week-day, organises his life to suit his climate; 
it is only at week-ends and on holidays that his actions are 
dominated by the weather (or the weather forecast). That is 
why I have called this book "Climate," and not "Weather," 
in Everyday Life. 

Since the needs of the people in all countries are largely 
decided by their climates, and since it is the province of industry 



1 2 INTRODUCTION 

and commerce to supply these needs, the business man is 
especially concerned. He must suit his activities — what he sells, 
when and where he sells it — to the prevailing weather in his 
markets. Hence a good deal of this book aims at helping the 
business man in his decisions. To take an absurdly extreme 
case, no business man would export fur coats to the Equator, 
or refrigerators to Greenland ! But the effect of climate goes 
much further than that. The manufacturer, having decided 
what he will make, has to decide where he will make it. Some 
products require special conditions, dry or damp air, freedom 
from soot or dust, a moderate or high temperature. Many 
industries are affected by rapid or large changes of temperature, 
especially the paint and varnish, electrical insulation and 
textile industries. The ease of manipulation of plastic materials 
is affected by temperature. Humidity affects printing; the size 
of paper changes with the relative humidity, causing loss of 
accuracy in multi-colour work; on the other hand, in very dry 
air friction in the presses produces static electricity which is 
a source of trouble. Dust is the enemy of all kinds of delicate 
machinery. These are only a few examples; the list is endless. 
In choosing the site of a factory all these climatic factors have 
to be balanced against each other and against considerations 
such as ground and building costs, labour, cost of transport, 
etc. 

The product having been made, it has to be packed. Some 
articles, such as cosmetics, confectionery and photographic 
supplies, are very sensitive to moisture, especially if the latter 
is combined with heat or rapid changes of temperature. If such 
goods are to be exported to a moist, tropical country, or even 
if they have to pass through such country on their way to 
their destination, special packing may be required. In certain 
climates air-tight packing is necessary. Canned goods may 
have to be coated with a protective varnish, or the labels affixed 
with waterproof gum. But such precautions are expensive, and 
are only to be taken if they are really necessary. Hence the 
weather to be expected in the market country and along the 
route to it must be studied to obtain sufficient protection at the 
smallest cost. Markets might be classified into three or four 
groups from this point of view, and different, suitable, packing 
provided for each group. Besides what may be termed the 
structural design of the packing, the artistic design (and the 



INTRODUCTION 1 3 

advertisements) may also vary from country to country accord- 
ing to the climate. 

Climate is important in other ways. A plentiful and reliable 
water supply is a necessity for civilised life, and all water supply 
depends ultimately on the rainfall, either locally or in some 
more or less remote gathering ground. Hence we must study 
the average rainfall, its seasonal distribution and its variability, 
especially the risk of drought. Dwellers on the banks of rivers 
are concerned with floods. Wind is important in all aspects of 
life ; our buildings must withstand wind pressure ; in the country 
we may depend on wind to pump our water and generate our 
electricity; in towns the wind carries away from house and 
factory chimney the smoke and gases which would otherwise 
stifle us. The sun is the greatest source of light and heat; these 
essentials can be manufactured, but at a price, the natural 
article costs nothing. The cost of a town fog in lighting alone 
is very great. We must be able to carry ourselves and our 
marketable products from one place to another; hence we must 
think of fog, snow, ice and other obstacles to transport. We 
need relaxation, and must care for the natural amenities of life. 
Finally, since accidents will happen in even the best-regulated 
climates, we must estimate the risks of blizzards, hurricanes, 
tornadoes, hail, lightning and other violent disturbers of the 
peace, and think how far we can guard against them. 

The purpose of the following chapters is to provide in a small 
compass answers to the questions which are most often asked 
about climate by anyone choosing a site, planning a house, 
factory or new town, starting up or expanding a business, 
buying an outfit for a journey overseas, or engaged in any other 
activity in which the prevailing weather conditions are an im- 
portant factor. To fulfil this ambitious programme completely 
would require a large encyclopaedia, but it is hoped that die 
references at the end will round out the outlines in the text. 
Most meteorologists have had the experience of providing, at 
considerable labour, answers to the wrong questions. It is 
hoped that at the least this book will help enquirers to ask the 
right questions, and so save the time of themselves and others. 

The general plan is as follows : Part I is an account of climate 
from the point of view of the man living in it. Chapter I, "The 
Economics of Climate," and Chapter II, "The Siting and 
Design of Houses and Factories in Relation to Climate," are 



14 INTRODUCTION 

intended as a primer on the exploitation of natural climatic 
resources. Chapters III to VII describe briefly living and 
working in the climates of different parts of the world. In 
Part II climate is regarded as an enemy to be faced. Chapter 
VIII describes recent researches into the relations between 
climate and the deterioration of various sorts of manufactured 
products. Chapter IX discusses the evil of dirt and dust in the 
air, and Chapter X all the various disasters from frost to tor- 
nadoes which afflict different parts of the world, where and 
how often they occur. Part III describes how to get over 
climate by heating and air conditioning, by lighting, by 
clothing, and by protection from the more violent and disastrous 
efforts of nature. 

Some people like figures, and Appendix I caters for them by 
giving statistical outlines of the climates of a number of places 
in all parts of the world. Some people do not like figures, and 
so each line of statistics is summed up in a brief thumb-nail 
sketch, based on precise definitions. Finally, since Babel still 
afflicts climatology, the last page gives a few necessary con- 
version factors between different systems of units. 



PART I 
LIVING WITH THE CLIMATE 



CHAPTER I 
THE ECONOMICS OF CLIMATE 

CLIMATIC DATA 

THE most important elements of weather to be considered 
in planning human activities are temperature, humidity, 
rainfall and wind. Good tables and maps of the distribu- 
tion of temperature and rainfall are available for nearly all 
parts of the world, and can be bought or seen at the offices of 
the national meteorological services. Humidity and wind are 
measured at many weather stations, but are less easily mapped, 
so that charts showing the distribution of these elements are less 
readily obtainable. Humidity can be represented in various 
ways : ( i ) by the absolute amount of water vapour in the air, 
measured in grams of water per cubic metre of air, or grains 
per cubic foot; (2) by the pressure of the water vapour in milli- 
bars; (3) by the "mixing ratio," or weight of water vapour per 
kilogram of dry air; (4) by the "dew point" or temperature at 
which the air would become saturated; (5) by the relative 
humidity, or percentage ratio of the amount of water vapour 
in the air to the amount which it would hold if saturated at the 
same temperature; (6) by the "saturation deficit" or the 
difference between the amount of water vapour and the 
amount which the air would hold if saturated; (7) by the "wet- 
bulb temperature," which is the temperature of a freely 
evaporating moist surface. Each of these measures is important 
for different purposes, as will appear in later parts of this book. 
A rough conversion table from one to another is given in 
Appendix II. For practical purposes, when exact data are not 
available, temperature and rainfall give a fairly good indication 
of the humidity. Other useful information concerns the fre- 
quency of rain, snow and fog, and the highest wind speed to be 
expected. In order not to overburden the text with tables most 
of the statistical climatic data are collected in Appendix I. 
World maps of air temperature for January and July are given 
in Figs. 2 and 3, and of wet-bulb temperature in Figs. 4 and 5. 
Classification of Climates. — For quick reference to die climate 



1 8 CLIMATE IN EVERYDAY LIFE 

of any particular district it is useful to have a map showing the 
distribution of different types of climate. Many such maps 
have been prepared by geographers, mainly on the basis of 
plant life. The main divisions shown on such maps are into 
regions classified as: Hot; Monsoon (cold dry winters, wet 
summers) ; Mediterranean (hot dry summers, cool wet winters) ; 
Desert; Temperate; Gold. This makes a good beginning, but 
for purposes of human activities it needs some modification. I 
have in Fig. I divided the world into nine regions. First we 
eliminate those parts of the world which have little interest for 
our purposes : 

( i ) Polar regions and tundras. These are inhabited if at all 
only by scattered hunting or fishing tribes ; their chief economic 
product is fur. A convenient boundary is the poleward limit of 
agriculture. 

(2) Mountain regions (general elevation above 6,000 feet). 
Apart from winter sports and holiday centres and mining 
districts, these generally have only a scattered population; 
access is difficult. Large elevated plateaus form an exception 
especially in the tropics, where such regions are important 
because of their healthfulness. 

(3) Deserts (annual rainfall below 10 inches). Apart from 
irrigated regions, such as the Nile Valley, and some mining 
centres, these also are very sparsely inhabited. 

The next three climatic regions, though often densely popu- 
lated, suffer from some climatic disadvantages. 

(4) Insolation climates, where the main feature of life is the 
intense solar radiation. At least once in two years the shade 
temperature exceeds no°F., and in some years it climbs to 
120 or more. Objects exposed to the sun rise to very high 
temperatures; 150 is not uncommon and 200 has been 
reached. These regions are rather dry and dusty in summer 
but generally have sufficient water for agriculture, either from 
winter rains or from irrigation. The high day temperatures 
cause trouble with any materials which soften with heat, such 
as photographic films, confectionery, electrical insulation, etc., 
and these troubles are aggravated by condensation at night, 
due to the very large daily range of temperature, and this causes 
powdered products to cake. These regions are often subject to 
strong winds carrying much fine penetrating dust. Activity is 
impossible during the hottest part of the day, and there is risk 



THE ECONOMICS OF CLIMATE 



19 




20 CLIMATE IN EVERYDAY LIFE 

of heat stroke. In spite of all these disadvantages they are often 
economically important, owing to the high crop yields under 
irrigation. Egypt is the outstanding example. 

(5) Deterioration climates. These are regions of high 
humidity combined with steady high temperatures which, 
however, are not extreme. Metals corrode rapidly and leather 
goods, clothes, paper, etc., soon go mouldy. The native popu- 
lations lack energy and initiative, and white immigrants cannot 
maintain their efficiency for many years without occasional 
recourse to a cooler, more stimulating climate. Tropical diseases 
are rampant unless special precautions are taken against them. 
Vegetation grows abundantly; and this zone includes the dense 
equatorial forests of Brazil and the Congo. Where the forests 
have been mastered these regions are of very great importance 
for the supply of tropical agricultural products. 

(6) Typhoon and tornado climates. These are regions, such 
as the West Indies, south-eastern U.S.A., parts of the coast 
of India and China and sub-equatorial islands, like the 
Philippines, having climates which are on the whole favour- 
able, but which are liable to destructive winds exceeding 
120 m.p.h. Although the intervals between these catastrophes 
are generally many years long, the risk is always present, 
and permanent buildings must be specially built to withstand 
them. 

(7) "Mediterranean" climates. These regions are very 
pleasant. The winter climate is ideal, but the summers are too 
hot for sustained energy. They are most extensive on the coasts 
of the Mediterranean Sea, whence they take their name, but 
they also occur in southern California and a few other parts of 
the world. They produce wine, olives and similar deep-rooted 
crops. 

(8) Cyclonic-temperate climates. The main feature of these 
climates is their changeability. The passage of cyclonic depres- 
sions maintains a continuous series of rapid changes of tem- 
perature in all seasons, with cloud, rain and wind alternating 
with spells of fine bright weather. Rainfall is generally sufficient 
at all seasons; persistent extreme heat or cold is rare. The 
climate is very stimulating, and the peoples of these regions are 
the most energetic and progressive in the world. The main 
regions included are parts of western and eastern U.S.A. and 
Canada, north-west Europe, Japan, parts of Chile, south- 



J 



THE ECONOMICS OF CLIMATE 21 

eastern Australia and New Zealand. The importance of these 
regions in the world's economy has been emphasised by 
Ellsworth Huntington (1924). 

(9) Winter Frost and Snow Climates. These include the 
interiors of North America and Eurasia in the same latitudes 
as the cyclonic climates of the coasts. Their main charac- 
teristic is the long cold winter, in which the ground is snow- 
covered for some months. In hilly regions the snow may be a 
serious obstacle to transport, but over flat country deep snow- 
falls are not very frequent. Nevertheless, in all the colder parts 
of this region it is necessary to provide against these occasions 
either by strengthening the roofs or by increasing their pitch. 
Dwelling houses and factories need thick or insulated walls and 
double windows to keep out the cold, and some form of central 
heating is required. Summers are rather warm, with a moderate 
amount of rain. 

Parts of Fig. 1 are left blank. These fall into none of the nine 
categories and show no special characteristics, either favourable 
or unfavourable. They mostly have a warm climate with 
pleasant, dry winters and a summer rainy season. 

Working Efficiency in different Climates. — Climate affects industry 
in two ways : by raising or lowering the general efficiency of the 
workers, and by its effect on the actual processes of manu- 
facture. The main climatic factors affecting general efficiency 
are temperature, humidity and wind. The human body has a 
normal temperature of 98 F., which is maintained against the 
loss of heat to the surrounding air by the "combustion" of food 
in the body. This combustion also provides the energy for all 
activity, both mental and physical. Mental activity uses up 
comparatively small amounts of energy, but hard physical work 
uses up a great deal. In the process the body generates heat, 
which has to be dissipated to enable the body to maintain its 
normal temperature. Excess of production over dissipation of 
heat results in a rise of body temperature, which causes dizzi- 
ness, a sense of depression, lassitude, and if it goes too far ends in 
cramps and heat-stroke. 

Heat is dissipated in two ways, first by conduction and 
radiation, and secondly by evaporation. Conduction depends 
on the temperature of the air and the wind speed, and on the 
amount of clothing worn. Radiation depends on the tempera- 
ture of the surrounding objects, which in a room means the 



22 CLIMATE IN EVERYDAY LIFE 

walls, ceiling and floor. A lightly clad office worker in a room 
in which the air is moist and free from draughts is most 
comfortable in a temperature of 60-65 ° F., i.e. the temperature 
at which the heat produced by the normal bodily activity just 
balances that lost by conduction and radiation. The introduc- 
tion of appreciable air movement, such as by opening a window, 
even if the outdoor temperature is the same, causes a feeling of 
chilliness because the rate of conduction is increased. When 
the air temperature is above 98 F. the effect of conduction is 
to add heat to the body instead of removing it, and this effect 
is greater the stronger the wind. Strong winds at very high 
temperatures, such as occur in desert storms, may add to the 
body more heat than it can dispose of by other means. The 
result is rising body temperature, ending in coma and death. 
Similarly, exposure to strong sunshine may cause the gain of 
heat by radiation to exceed the loss, and to counterbalance that 
the loss of heat by conduction must be increased, either by 
lowering the temperature of the air or by increasing the ven- 
tilation. If the solar radiation cannot be compensated in this 
way heat-stroke results. 

Heat-stroke is naturally most frequent in the tropics, both in 
regions of very high temperature such as the Punjab, Sind and 
North-west Provinces of India, Iraq and the dry, hot parts of 
Africa and Australia, and in places of damp heat such as the 
Persian Gulf, the west coast of Africa, Burma and Malaya. 
Heat-stroke also occurs, however, in "temperate" regions 
during abnormally hot summers, especially in the northern 
United States and even in Canada and Europe. Some people 
are more liable to heat-stroke than others; this can be deter- 
mined by simple tests. 

The loss of heat by evaporation depends on the amount of 
moisture in the air. Even when the body is cool there is always 
some evaporation in breathing, which is made visible in steam- 
ing breath in frosty weather. Since the air coming from the 
lungs is warmed to body temperature and saturated, there is 
always some loss of heat by evaporation in this way, the amount 
depending on the temperature and humidity of the air. 

When, owing to high air temperature, sunshine, heavy 
manual work or other causes the body produces more heat than 
can be disposed of by conduction, radiation and evaporation 
from the lungs, the sweat mechanism comes into operation. 



THE ECONOMICS OF CLIMATE 23 

The body becomes wet, and so long as the air is not saturated 
and can reach the body, active evaporation takes place. Under 
similar conditions of clothing the rate of evaporation depends 
entirely on the wind speed and the saturation deficit, or the 
difference between the amount of water vapour in the air before 
it reaches the skin and the amount which it can hold. If the air 
is dry, evaporation goes on briskly and the body temperature 
remains normal. But if the air is hot and nearly saturated it 
cannot evaporate water so well, and consequently its cooling 
power is small or even zero. Under such conditions continual 
heavy work becomes impossible. A rough upper limit of com- 
fortable conditions when the wind speed is small is given by the 
following table : — 



Temperature, ° F. . 


68 


70 


75 


80 


85 


90 


95 


Humidity per cent. 


80 


77 


69 


58 


46 


35 


25 



Above these limits the conditions are "sultry" at the higher 
humidities and hot and irritating at the lower humidities. 

A good measure of the cooling power of the air by conduction 
and evaporation combined is given by the wet-bulb thermo- 
meter. This is an ordinary thermometer in which the bulb is 
covered by a layer of muslin kept moist by a wick dipping into 
a container of distilled water. When the air is saturated there 
is no evaporation and the thermometer gives the air tempera- 
ture. When the air is not saturated evaporation depresses the 
temperature of the wet-bulb below the air temperature, and 
the difference is a measure of the dryness of the air. 

Hard physical work even with light clothing becomes very 
difficult when the wet-bulb temperature rises above 85 ° F., and 
is practically impossible with a wet-bulb above 90 F. Light 
work is possible up to a wet-bulb of 88° F. in still air or 93 ° F. 
in a moderate breeze; 78 F. has been taken as the limit for 
white settlers in the open, but experience during the war has 
shown that this limit is too low. For half-naked men the limit 
can be raised by 1 -2 F. Figs. 2 and 3 show the average air 
temperature at sea level in January and July, and Figs. 4 
and 5 show the average wet-bulb temperatures (approximate) 
in the same months over the land areas, not reduced to 
sea-level. 

The air temperature at any height can be found approxi- 
mately by subtracting from the sea-level value an amount of 



24 



CLIMATE IN EVERYDAY LIFE 




THE ECONOMICS OF CLIMATE 



25 




26 CLIMATE IN EVERYDAY LIFE 

3° F. per thousand feet of altitude. For wet-bulb temperatures 
this relation does not hold; they decrease more slowly at high 
than at low temperatures. Below 32 F. the wet-bulb differs 
very little from the dry-bulb temperature. 

The maps of mean wet-bulb temperature (Figs. 4 and 5) are 
based on the mean air temperature and mean relative humidity 
for the twenty-four hours at about a thousand well-distributed 
stations. The mean wet-bulb calculated in this way is slightly 
higher than that calculated directly from hourly readings of the 
wet-bulb thermometer, but some experiments showed that the 
error is small — rarely as much as i° F. — and the maps can be 
accepted as substantially accurate except in regions of rugged 
topography, where it was impossible to show the effect of relief. 
The diurnal variation of wet-bulb temperature is much smaller 
than that of the air temperature, especially in hot regions, and 
is generally about 6° F. in middle and low latitudes. Conse- 
quently the wet-bulb during the hottest hours of the day can 
be estimated by adding 3 F. to the reading on the chart. Since 
a wet-bulb temperature of 78 F. is often quoted as about the 
limit for permanent settlement by whites, the wet-bulb isotherm 
of 75 F. is an important climatic boundary, and this is shown 
by a heavy line. 

In January the isotherms of 75 ° F. include only two regions 
along the Equator, one from Brazil to the Guinea Coast of 
Africa, and the other from East Africa across the Indian and 
Pacific Oceans to about 155 E., including northern Australia. 
In this month the wet-bulb temperature nowhere exceeds 
80 ° F. In July there are several small areas above 75 ° F. 
between the Gulf Coast of U.S.A. and northern Brazil, 
two small areas in West Africa, one over the Red Sea, and 
a long stretch from the Persian Gulf across India, Indo-China 
and southern China to 155 E., which includes two small 
areas with a wet-bulb above 8o° F. There are two "islands" 
below 75 F. over the high ground in the interior of India 
and Indo-China. 

With a mean wet-bulb above 70 F. conditions are likely to 
be uncomfortable on some afternoons. The most comfortable 
conditions for Europeans will be found between the wet-bulb 
isotherms of 50 and 6o° F. 

Although the effect of topography in mountain regions could 
not be shown on the maps, the lower wet-bulb temperatures at 



THE ECONOMICS OF CLIMATE 27 

high levels in the tropics are important for health stations. A 
few figures are given below: — 

Table i . — Variation of wet-bulb temperature with height. 





Height 


Shade Temperature 


Mean Wet-Bulb 




feet 


Jan. 


July 


Jan. 


July 






°F. 


°F. 


°F. 


°F. 


India — 












Lahore 


702 


55 


90 


49 


80 


Simla . 


7>28 3 


40 


65 


35 


62 


Ceylon — 












Colombo 


24 


79 


81 


79 


77 


Diyatalawa 


4,101 


65 


70 


62 


63 


Nuwara Eliya 


6,170 


57 


60 


54 


58 


Malaya — • 












Bukit Mertajam . 


65 


80 


81 


74 


76 


Cameron Highlands 


5,120 


64 


65 


61 


62 


Gunong Tahan 


5,460 


60 


64 


58 


60 


Java— _ 












Batavia 


26 


78 


79 


75 


74 


Tosari 


5>677 


62 


59 


59 


56 


Switzerland — 












Zurich 


1,617 


32 


64 


30 


59 


Davos 


5,118 


19 


54 


18 


5i 


Santis . 


8,200 


17 


42 


15 


40 



The highest wet-bulb temperatures probably occur in the 
Red Sea and Sierra Leone areas, in both of which reliable 
readings have been known to exceed 90 F. Readings of ioo° F. 
have been quoted, but are doubtful. In Britain a wet-bulb 
temperature of 70 F. may be expected once in two years ; the 
highest reading in thirty-four years was 76 F. 

A more exact measure of the cooling power of the air 
in different circumstances is given by Dr. Leonard Hill's 
"katathermometer." The bulb is dipped in hot water at 
a temperature of just over ioo° F. and the time taken for 
its temperature to fall from 100 to 95 F. is measured. This is 
divided by a factor supplied by the manufacturer to give the 
cooling power of the air. The effect of clothing can be simu- 
lated by slipping a wet muslin glove over the bulb. This 
instrument is suitable for comparing conditions in different 
parts of a factory, but the readings are very local, and even if 
the observations were available they could not be made the 
basis of a world map. 



28 



CLIMATE IN EVERYDAY LIFE 




THE ECONOMICS OF CLIMATE 



29 




30 CLIMATE IN EVERYDAY LIFE 

When the air temperature is very high and the air very dry, 
as happens in sub-tropical climates in summer, the amount of 
evaporation is very great. The body sweats profusely in the 
attempt to keep cool, and this loss of moisture has to be made 
good by drinking an equivalent quantity of fluid. Since the 
sweat contains salts, these also have to be replaced, and the best 
drink is slightly salted water. Drinking only plain water is 
liable to cause cramp. The stomach can only absorb a little 
under two quarts of water an hour, and cooling by sweating 
cannot for long exceed that produced by the evaporation of 
this quantity of water. 

In offices and factories the effect of high temperature and 
humidity is accentuated by the heat and moisture produced by 
the workers themselves. Ventilation is only a partial remedy, 
for under extreme conditions the current of air required to give 
sufficient cooling would be so strong that it would raise dust 
and cause other inconveniences. The cooling power of the air 
increases only as the square root of the air speed, whereas the 
lifting of dust increases as the square of the speed. Air currents 
exceeding 500 feet per minute are impracticable for this reason. 
Under such conditions the only remedy is air conditioning. 

This limitation of the amount of possible physical work by 
the cooling power of the air is only one of the reasons why the 
output of manual workers in the tropics is less than in cooler 
regions. The other reason is the monotony of the climate. 
Tropical uplands may have an ideal temperature and humidity 
for comfort, but if the weather is the same day after day, effi- 
ciency falls off and is replaced by boredom and lassitude. 
Moreover, the body loses its power of resistance to small changes 
of temperature and becomes subject to chills. The best climate 
for efficiency is one in which there is continual variety — spells 
of fine weather interrupted by the passage of storms, with 
corresponding changes of temperature and humidity. The best 
climate for all-round activity, physical and mental, is found in 
western Europe, the north-eastern United States and neigh- 
bouring parts of Canada, the coast of California, south-east 
Australia, Tasmania and New Zealand. Probably south-east 
England has the finest climate in the world from this point of 
view. Too much storminess goes to the other extreme, monotony 
resulting from the absence of relief by fine spells; continuous 
cloudiness is as depressing as continuous hot weather. The 



THE ECONOMICS OF CLIMATE 3 1 

whole temperate region between latitudes 30 and 6o° is 
moderately favourable, except in the far interior of Asia. The 
Arctic regions are unfavourable because of the numbing effects 
of extreme winter cold, man's whole energy being taken up in 
keeping warm. Monsoon countries such as India and China 
gain from the marked alternation between the seasons, the dry, 
cool winter helping the inhabitants to carry on through the hot 
season, while the summer rains again bring relief. Even in 
countries most favoured climatically there are sometimes long 
periods of drought and high temperature, and these have been 
found to be associated with periods of industrial depression. 

The loss of efficiency caused by climate can be minimised in 
various ways, such as the siting of houses and factories to take 
advantage of prevailing winds (Chapter II) or by air con- 
ditioning (Chapter XI). But apart from that, much can be done 
to provide amenities. Regular physical exercise at a suitable 
time of day (not the hottest part) is a partial substitute for 
variability of climate in maintaining efficiency, and mental 
variety is also helpful. In hot climates subject to much strong 
sunshine, loose white clothing is desirable. 

After some time in a different climate the body tends to 
become acclimatised, especially after migrating from a cooler 
to a warmer climate, when the problem is that of disposing 
more readily of surplus heat. Migrants from a warmer to a 
cooler climate on the other hand have to produce more heat to 
counterbalance the increased loss, and this causes exhaustion 
and greater liability to disease. For this reason it may not be 
economical to import labour from lower latitudes. 

SUITABILITY OF CLIMATE FOR SPECIAL INDUSTRIES 

It is well known that some manufacturing processes require 
special meteorological conditions, especially of temperature and 
humidity. The cotton industry, for example, requires rather 
moist air at a moderate equable temperature if the threads are 
to adhere properly. The most favourable conditions for spinning 
are said to be a temperature of 70-75 F. and a relative humidity 
of 65 per cent. ; for weaving the best humidity is about 75 per 
cent. The moist, equable climate of Lancashire provides 
favourable conditions, and this helps to account for the early 
development of the cotton industry there. Spinning and weaving 



32 CLIMATE IN EVERYDAY LIFE 

of wool requires a lower humidity, about 60-70 per cent., and 
this industry grew up in Yorkshire, on the drier side of England. 
Of course, geographical conditions also played a part, the port 
of Liverpool being favourably situated for trade with the cotton- 
growing regions of America. Spinning artificial silk is also said 
to require a high humidity. 

The manufacture of cigars and cigarettes requires only 
average conditions, but their conditioning calls for high tem- 
perature, around 90 F., and very moist air. At the other 
extreme is the manufacture of confectionery, which needs cool, 
dry rooms around 6o° F. and a humidity of about 50 per cent. 
These and, in fact, all edible products also need clean air, free 
from dirt and pollution. 

These are only a few examples selected at random; modern 
manufacturing processes are so complex that from start to finish 
a wide range of conditions is generally called for, and these can 
only be provided artificially. Moreover, the theoretically best 
conditions for the product may not be conducive to sustained 
output by the workers. In such cases a compromise is necessary. 
Generally speaking, a suitable outdoor climate and surroundings, 
with their good effect on the energy and efficiency of the workers, 
are more important than the requirements of the process rooms, 
since conditions in the latter can, within limits, be adjusted by 
air conditioning. The question of humidity in textile factories 
is especially important because the high humidity required for 
good products is very unfavourable to the welfare of the workers. 
Strict control is necessary and the conditioned air must be 
evenly distributed through the factory by proper circulation. 
Chen-li (1938), after investigating conditions in a cotton mill 
in China, advised that a better regulated humidification might 
double the efficiency of the weavers. 



POWER 

There are two important sources of power which depend 
directly on meteorological elements, namely wind and water; 
a third possible source is solar heat. 

Water power. — Running water has been used since very early 
times as a small local source of power to drive water-mills, the 
economic value of which was recognised in surveys such as 
Domesday. The use of water power in early times was, how- 



THE ECONOMICS OF CLIMATE 33 

ever, rather inefficient, and was limited by the fact that the 
natural conditions which are most suitable — rugged topo- 
graphy and rapidly flowing rivers — are not those which favour 
agriculture. The development of hydro-electricity, by enabling 
power to be transmitted to a distance, has removed this limita- 
tion. 

The value of a river as a source of power depends on three 
factors, the volume of water, the reliability of the flow, and the 
slope of the valley, especially the presence of waterfalls. The 
volume of water depends on the area of the gathering ground, 
the depth of the rainfall, and the proportion of the latter which 
reaches the river. One inch of rain over a square mile is equal 
to 2,323,000 cubic feet of water, but the losses by evaporation, 
transpiration and deep percolation may account for anything 
from one- tenth to the whole of this. Rainfall is greatest on high 
ground, and in mountain areas the proportion of run-off to 
rainfall is generally greatest, hence in many respects mountain 
districts provide the most favourable opportunities for de- 
veloping hydro-electric power. 

Reliability of rainfall depends on four factors: the way in 
which the fall is distributed through the year, the variability of 
the fall from year to year, the size of the gathering ground, and 
the storage of water in the sub-soil and in swamps and lakes. 
The variability of rainfall both through the year and from year 
to year is least in the rainier parts of the cyclonic-temperate 
regions, particularly in those countries where high ground 
fronts prevailing winds blowing off the sea. These regions have 
large annual averages of rainfall, so that there is plenty of water 
at all seasons. On the other hand, the gathering grounds are 
often small, since the rivers flow direct to the sea. Both lakes 
and waterfalls are most frequent in mountain districts which 
formerly bore glaciers, such as Scotland, Wales, Norway, the 
Alps and much of northern North America, and these mostly 
coincide with the "cyclonic temperate" climate. Where lakes 
and waterfalls are absent, it is necessary to construct artificial 
ones by building dams. 

The least favourable type of climate for hydro-electric works 
is, apart from deserts, the "Mediterranean" type, with its long, 
dry summer, in which the rivers fall to very low levels before 
they are replenished by the winter rains. Very cold climates 
are also not suitable, because the rivers freeze in winter. 



34 CLIMATE IN EVERYDAY LIFE 

Monsoon climates are somewhat better because the dry season 
comes in winter when evaporation is least. 

Rugged topography is necessary to provide a head of water 
to drive the turbines. The most favourable site is naturally a 
high waterfall in a large, steady river. The combination of a 
large river, high falls and the "Great Lakes" to equalise the 
flow, all in a locality suitable for extensive industry, makes the 
Niagara Falls highly favourable for the development of hydro- 
electric power on a very large scale. Other suitable regions are 
the Victoria Falls on the Zambesi, and the Nile where it issues 
from Lakes Victoria and Albert, but these African sites are all 
distant from industrial centres and the power will have to be 
transmitted over long distances. Small mountain valleys, such 
as those of western North America, cannot provide power on 
the same scale, but are useful as cheap local sources. 

Wind power, — Wind may be used as a source of power for 
pumping, generation of electricity, etc., as well as for ventila- 
tion and removal of noxious fumes. In most countries the 
variability of wind speed is too great for wind to be used as a 
source of power without either expensive storage batteries or an 
alternative source of power; hence it may not be economical. 

The power yield is proportional to the cube of the wind speed, 
so that with a wind of 30 m.p.h. it is twenty-seven times the 
yield at 10 m.p.h. At low speeds the yield is negligible, and 
modern wind generators do not become effective until the speed 
rises to about 8 m.p.h. On the other hand, the wind pressure on 
the sails becomes dangerous when the speed exceeds 30 m.p.h., 
and they are designed to turn into the wind and go out of 
action at about this speed. Hence the useful range of wind 
speed is between 8 and 30 m.p.h. (in India, where the winds 
are generally lighter than in Britain, the type of wind generator 
advocated becomes effective at 6 m.p.h., but has a correspond- 
ingly lower maximum; in the U.S.A. the effective minimum is 
regarded as 10 m.p.h.). 

The average wind velocity increases with height above the 
ground. The rate of increase depends on the nature of the 
surface, being least above water surfaces or smooth grassland, 
and greatest above broken irregular ground or surfaces broken 
up by buildings. It is also greatest at night and in cold weather, 
and least on hot sunny afternoons. As a general average in 
temperate regions we may use the following table, which gives 



THE ECONOMICS OF CLIMATE 35 

the speed at any height as a ratio to that at 33 feet (10 metres). 
This is the standard height of anemometers at meteorological 
stations from which most official statistics are derived. 

Table 2. — Variation of wind with height. 

Height (feet) . 5 10 15 20 30 40 50 60 100 150 200 300 

Ratio to 33 feet . . 73 -8i -875 -92 -99 1-03 1-07 i-i 1-21 1-29 1-35 1-46 

The power developed varies somewhat with the type of wind- 
mill. A trial French integrator was graded to give 0-37 (V/10) 3 
kilowatts per square metre of sail surface, V being the average 
wind velocity in metres per second. When V is expressed in 
miles per hour this is equivalent to 0-31 (V/10) 3 kw. per 100 
square feet of sail surface. The actual yield would probably be 
slightly less than this. 

Since the power developed is proportional to the cube of the 
wind speed, and the latter is roughly proportional to the fifth 
root of the height above the ground, the effectiveness of a 
generator is more or less proportional to the square root of the 
height of the sails above the ground. The increase of effectiveness 
is probably greater than this, however, almost linear, because 
the gustiness of the wind decreases with height. The most 
favourable site is on the crest of a long, high ridge lying across 
the direction of the prevailing wind. Between 100 and 200 feet 
above the crest the wind speed exceeds that at the same level 
in the free air above level country, the acceleration factor being 
the greater, the higher and steeper the ridge. A windward 
slope of 1 in 10 increases the wind speed over the summit by 
up to 15 per cent., of 1 in 7 by up to 23 per cent., and of 1 in 4 
by up to 35 per cent. Sites over very steep slopes are unfavour- 
able because the wind is thrown up nearly vertically and 
becomes very turbulent. 

A high average wind speed is not necessarily an advantage 
however. Owing to the variability of speed there are times 
when the speed is so high that the windmill shuts down, and 
other times when it is below the effective limit. With increasing 
average speed the duration of winds below, say, 8 m.p.h. 
decreases, and that of winds above, say, 30 m.p.h. increases, 
but the two changes do not cancel out. The net result is that, 
with winds of average steadiness such as are found in the British 
Isles, the duration of winds between 8 and 30 m.p.h. is greatest 
where the average wind speed is between 14 and 18 m.p.h. 



36 CLIMATE IN EVERYDAY LIFE 

The variation of effective duration with average wind speed is 
shown by the full curve of Fig. 6, which was calculated by a 
method devised by C. E. P. Brooks (1949). 

Table 3 gives for a few places the percentage time during 
which the observed wind was between 8 and 30 m.p.h. at 
33 feet, and the estimated time at 100 feet, for the year as a 
whole and in winter and summer. 



Table 3. — Percentage of time with winds between 8 and 

30 m.p.h. 













33 feet 




100 feet 




Lat. N. 


Long. 


Height 


Mean 
Speed 




























feet 


m.p.h. 


Win- 
ter 


Sum- 
mer 


Year 


Win- 
ter 


Sum- 
mer 


Year 


Tiree (W. Scot- 




















land) . 


56 32 


6 55 W. 


75 


15*9 


68 


67 


70 


62 


70 


66 


Abbotsinch . 


55 52 


4 26 W. 


65 


8-3 


44 


40 


44 


50 


52 


51 


Cranwell 


53 22 


1 37 W. 


284 


103 


68 


6l 


54 


67 


68 


69 


Felixstowe 


5i 58 


7 E. 


60 


I I '2 


65 


59 


61 


68 


66 


67 


Croydon 


51 21 


7W. 


3 X 3 


II O 


61 


46 


53 


67 


58 


62 


Calshot. 


50 49 


1 18 W. 


58 


12-9 


7i 


65 


68 


73 


73 


72 


Boscombe Down 


51 10 


1 45 w. 


462 


io-o 


64 


44 


64 


68 


53 


6l 


Manchester 






















(Barton) . 


53 28 


2 23 w. 


153 


9-i 


60 


46 


53 


64 


55 


60 


Aldergrove 






















(N. Ireland) 


54 39 


6 13 W. 


328 


100 


60 


48 


54 


66 


59 


63 


Harpenden . 


51 49 


21 w. 


— 


9-0 


61 


25 


49 


65 


34 


58 


Perpignan 


42 41 


2 55 E. 


— 


14-6 


— 


— 


4i 


— 


— 


38 


Brest . 


48 25 


4 30 W. 




15-8 






76 






72 



The first nine places are from averages for several years 
compiled by the Meteorological Office, London. The figures 
for Harpenden are for one year only, based on a Bulletin of the 
Institution of Agricultural Engineering, Oxford (1926). The 
figures for Perpignan and Brest are read off a diagram by 
P. Ailleret (1946). 

As the power yield is proportional to the cube of the wind 
speed, the greater part of the power is given by winds near the 
upper working limit, i.e. between 25 and 30 m.p.h. These are 
more frequent with high than with low average speeds. The 
power yield of a windmill operating between 8 and 30 m.p.h. 
for different average speeds is shown by the broken curve of 



THE ECONOMICS OF CLIMATE 



37 



Fig. 6, which indicates that the greatest power yield is given 
by an average speed of about 20 m.p.h. The scale on the right 
gives the power yield as a percentage of that of a windmill 
running full time at 30 m.p.h. Since reliability is desirable 
as well as maximum yield over short periods, the optimum 
average wind speed is probably about 18 m.p.h. This is 
very little above the average wind speed at 33 feet in the 



70 



60 

J 50 

■J 40 

s_ 
§-30 

cn 



10 



























r *s 














/ 


X i 


V 












/ 




\ 












/ 




\ 












/ 




\ 








J 




1 




\ 


V 






M 


/ 














•§/ 


/) 








X 






a 


! 


f 








1 




$1 


1$ 














7 


1$ 














/ 


1 














/ 1 
















/ / 
















f 1 
















1 














1 
















1 














/ 


1 














/ 


t 














/ 


1 














/ 


t 















2-1 



1-8 



1-5 2 



1-2 



0-9 



O 

a_ 

c 
o 



0-6 



0-3> 



30 



O 
40 



o to ao 

Mean wind velocity m.p.h. 

Fig. 6. — Operating time and power yield of a windmill at different 
mean wind speeds. 



windiest parts of Britain, and in such places there would be 
little advantage in seeking greater wind speeds eitiier by raising 
the sails to great heights above the ground or by going to the 
windiest ridges. Inland, however, where the average wind 
speed is much less than on the western coasts, a lofty installation 
would be an advantage. 

At places where the wind is steady, such as islands in the 
Trade winds, the curves of Fig. 6 are much steeper, so that if the 
average wind speed is between 15 and 20 m.p.h. a windmill 



38 CLIMATE IN EVERYDAY LIFE 

operating between 8 and 30 m.p.h. will rarely be idle. If the 
average wind speed is below about 12 m.p.h. a windmill operat- 
ing between lower limits of wind speed would be more reliable. 
The Use of Solar heat as a source of power. — The suggestion has 
often been made that the heat of the sun's rays could be used 
as a source of power to work a heat engine. It is possible that 
in a sub-tropical country with almost continuously clear skies, 
such as Egypt, this source of power might be economical, but 
in general the yield of power is low for the area of plant re- 
quired, and it is doubtful whether, even in the most favourable 
circumstances, the value of the yield expressed as a percentage 
of the cost of installation can compare with the cost of other 
sources of power. A discussion of the possibilities is given by 
H. C. Hottel (1941) and F. Trombe (1948). 



WATER SUPPLY 

Most large towns have a reasonably well-assured supply of 
water. Where much water is used, and has to be obtained from 
local sources, the possibility of drought has to be considered. 
The practice of water engineers in Britain is to provide sufficient 
storage to ensure the water supply against the three driest con- 
secutive years to be expected in an average period of fifty years. 

The sources of water supply may be divided into rainfall; 
wells and springs; rivers, lakes and reservoirs. 

( 1 ) Rainfall. — In this country few areas depend solely on the 
artificial storage of rainfall, but in other parts of the world this 
may be the only source of supply. Even in Britain rain may 
serve as an auxiliary source, or for processes for v/hich soft water 
is essential. The gathering ground consists of the roofs of all 
the buildings, whence the water is led by pipes and gutters into 
covered cisterns. When the cisterns are full any further rain is 
not available for use. Hence there are two problems to be 
faced : first, the provision of sufficient storage capacity to tide 
over the longest dry period to be expected; secondly, the pro- 
vision of a sufficient gathering area to keep the cisterns full 
during periods of rain. 

The period of drought to be provided against must be cal- 
culated from records of daily rainfall at the site or at some 
neighbouring comparable place. These records can be obtained 
from the official weather service of the country concerned, or 



THE ECONOMICS OF CLIMATE 39 

through the British Meteorological Office. For this purpose 
drought must be defined, not as the period entirely without 
rain, but as that between the last fall sufficient to top-up the 
cisterns to capacity and the first fall sufficient to wet the roofs 
and leave something over. It must be remembered that the 
whole of the rain does not reach the cisterns ; some is evaporated 
from the roofs, the amount lost being greater if the roofs are 
covered by slightly pervious tiles than from impervious slates 



r*"i5 




Q | JAN. ) FEB. , MAR. , APR. ■ MAY .JUNE , JULY , AUG i SEP , OCT , NOV. , PEC. I Q 

Fig. 7. — Level of a rain-storage tank, London, 1938. 

or galvanised iron. An allowance of 10 per cent, should be 
made for this loss. 

In regions with an uncertain rainfall or with a long, dry 
season, provision against drought will be the governing factor; 
in wetter regions with a more regular rainfall the daily require- 
ment. The safest procedure is to select from the records the 
driest year and plot the water-level in the cistern day by day. 
Suppose that the requirement averages X gallons a day, which 
is equivalent to a rainfall of x inches on the roof-space available; 
x might be, say, 0-04. We will also suppose that on each rain- 
day the first o-o 1 inch in winter and the first 0-02 inch in sum- 
mer is lost by evaporation from the roofs and gutters without 
reaching the cisterns. Let the capacity of the cistern be T 
gallons, equivalent to a rainfall ofjy inches \y might be 5-0 (three 
months' supply). We can then proceed to calculate the level of 
water in the cistern day by day. 

An example is shown in Fig. 7, which represents the variation 



40 CLIMATE IN EVERYDAY LIFE 

of level of a cistern under these conditions in London during the 
dry year 1938. Since we are starting in winter, when the rain- 
fall is generally sufficient, we will start with the cistern at a 
level equivalent to a rainfall of 4-5 inches. 1st January was dry, 
so the level drops to 4-46. On 2nd January the rainfall was 
0-08 inch, so the level rises by 0-08 inch less o-oi inch evapora- 
tion and 0-04 inch consumption, i.e. by 0-03 inch to 4-49 inches. 
After some fluctuations the level reaches 5 inches, and any 
further rainfall runs to waste. During the dry spring and 
summer the level falls steadily and by 6th August it is down to 
0-18 inch, which is definitely in the danger zone, but after that 
wetter weather sets in and by mid-December the cistern is full. 

Allowing for a 10 per cent, loss, each gallon of water a day 
requires about 800 /R square feet (horizontal projection) of roof, 
where R is the lowest twelve-monthly rainfall to be expected. 
In London, for example, the rainfall in twelve consecutive 
months is not likely to fall below 11 inches, so that a daily 
requirement of 100 gallons would be assured by a horizontal 
roof area of 7,300 square feet. Calculations along these lines 
will ensure against risk of water shortage, and may also avoid 
the expense of providing unnecessarily large storage. 

(2) In wells and springs natural storage takes the place of 
artificial cisterns, and generally provides a greater reserve. 
The yield of large wells varies from a few hundred thousand to 
two million gallons a day. These natural sources draw their 
supplies from a larger area than roofs, but losses by evaporation 
and run-off are also greater. All the factors being uncertain, 
the question of reliability can only be determined by past 
records of yield or, if these are not available, by a geological 
survey. 

Wells are generally most reliable on low ground near springs 
or streams, and least reliable on high ground. Since bulk 
storage of water from wells and springs is rarely practicable, 
the reliable yield is the actual minimum observed or calculated 
over a considerable period of years. It should also be remarked 
that the pumping of water from a well lowers the water-table 
in the rocks, and hence the yield of the well. Doubling the 
number of wells in a small area does not usually double the 
yield; in fact the increased supply may be quite small. Accord- 
ing to H. Lapworth (1930), the effect of a large well on the 
water-table may extend to a distance of one or two miles; a well 



THE ECONOMICS OF CLIMATE 4 1 

in California was even shown to affect the water-table at a 
distance of 5 miles. 

(3) Rivers, lakes and large reservoirs are the most reliable 
source of supply. Sufficient information as to the levels of rivers 
in past years is generally available to serve as a guide, but 
riparian rights have to be considered. The amount of water 
taken from a river must not be so great as to interfere with the 
users further downstream, and if the used water is returned to 
the river it will probably have to be purified first. Lakes and 
large reservoirs generally provide sufficient reserves to cover 
all reasonable vicissitudes, but they are liable to other troubles, 
such as silting or the growth of organisms. 

The usual practice in estimating the reliability of a natural 
source of water dependent on rainfall is to take as a basis the 
smallest annual rainfall to be expected in three consecutive 
years during a period of fifty years, the losses being regarded as 
constant. In the British Isles this is usually taken as four-fifths 
of the average annual rainfall; it can be estimated more 
accurately from a comparatively short period of observations 
by statistical methods. If the average annual rainfall over a 
period of N years is R inches, the first step is to write down the 
difference between the rainfall of each individual year and the 
average R. These are added up, ignoring minus signs, and 
divided by the number of years N. This gives the mean devia- 
tion D, or the average amount by which the rainfall of any year 
differs from the average rainfall. Then the smallest rainfall 
to be expected in any one year out of fifty is less than R by 
about two and a half times D. The smallest rainfall to be 
expected out of fifty sets of three consecutive years each is less 
than three times the average annual rainfall by about four and 
a half times D, or, allowing for some persistence of drought 
from year to year, smallest three-year rainfall equals 3R— 5D. 

The loss by evaporation is not constant from one year to the 
next, but varies with the rainfall and temperature. In dry 
summers the soil dries out and consequently the total loss by 
evaporation decreases with decreasing rainfall, but less rapidly. 
Evaporation increases with increasing temperature, by from 
0*6 to 1 inch a year for each increase of mean annual tem- 
perature by 1 ° F. As, in Britain, rainy summers are generally 
cool, the two factors more or less cancel out and the observed 
loss by evaporation remains fairly steady at from 11 to 17 inches 



42 CLIMATE IN EVERYDAY LIFE 

a year, increasing from north to south (H. Lapworth, 1930). 
Evaporation is almost confined to the summer half-year, and 
in Britain rivers most often reach their lowest level of the year 
in September. 

The disposal of waste water may be a serious problem. The 
subsoil is generally an excellent natural filter (apart from the 
presence of cracks) , and in dry country with porous rocks a sump 
well away from the source of water supply is the best means of 
disposal. If the ground becomes waterlogged or frozen this is 
no longer practicable. Meteorological records will show the risk 
of frost. Waterlogging depends mainly on the nature of the 
subsoil and partly on the difference between the rainfall and 
evaporation at different seasons. The risk can best be diagnosed 
by the study of past records ; levels of water in wells are useful 
for this purpose. 

TRANSPORT 

The last aspect of the economics of climate to be mentioned 
is the possibility of interference of access by fog, snow, ice, floods 
and similar obstacles. Fog also adds to the cost of lighting. 
Fog consists of minute water droplets suspended in the air. It 
is most frequent in humid climates, especially where the rainfall 
is moderate. There are two types of fog: radiation fog and 
advection fog. Radiation fog occurs on low ground, especially 
broad, level river valleys, when the temperature falls owing to 
radiation on cold nights; it is caused by the air cooling below 
the dew point. Flat river valleys such as lower Thames and 
Lea valleys near London are especially liable to fog because 
the air is nearly but not quite stagnant and there is a good 
deal of water about to keep the air moist. 

Advection fog is caused by the cooling of air blowing over a 
colder surface or by the mixture of two moist air currents at 
different temperatures. It is especially common where a cold 
current flows along a coast, as in California and north-eastern 
North America. 

Fog is especially prevalent over large towns. This is partly 
because the chemical particles in the smoky air provide 
nuclei for the condensation of moisture, and partly because the 
smoke particles themselves decrease the visibility. Where 
country fog is white and clean, town fog is black and dirty. In 
addition to the London region, town fogs are very frequent 



THE ECONOMICS OF CLIMATE 43 

along the coastal regions of Belgium, Holland and north-west 
Germany. 

Owing to the local nature of fog it is not possible to give 
a world map of its distribution. The official international 
definition of a fog is a horizontal visibility of less than 1,100 
yards (1,000 metres), but some of the figures may refer to 
different criteria. Visibility of this range interferes with flying, 
but has little effect on land transport. "Thick fog," defined as 
visibility less than 220 yards, interferes with all forms of traffic; 
on the average it occurs on five or six days in each winter in 
London. In the United States before 1940, thick fog was 
defined as visibility less than 1,000 feet; with light fog it might 
be 6 miles or more. In the centre of large towns such as London 
the warmth of the town itself dissipates the fog at ground level, 
and a fog often takes the form of a dense overhead cloud not 
much above roof level, which produces the darkness of night 
in the streets. Lights have to be lit in all buildings, but visi- 
bility remains quite good at street level. 

The worst fogs can generally be avoided by selecting a site 
on the lower hill slopes above the valleys, but not on the tops 
of the higher hills. In the London area, for example, there are 
two belts, one of radiation fog in the valleys and the other, which 
really consists of low cloud, on the hills above a height of 500- 
600 feet. Free air drainage is essential, and in consequence the 
least foggy situations in any district are also those most free 
from frost. 

Snow. — Places in cold climates record anything up to 100 
days with snow a year. Many of these are light falls which 
soon melt, but others are heavy falls, often accompanied by 
strong winds which pile up the snow in deep drifts and cause 
widespread interruption of road and rail transport. These 
include the famous "blizzards" of North America, but similar 
conditions also occur in this country and in Europe and Siberia. 
An account of these blizzards is given in Chapter X. 

The best criterion of the effect of snow on human activities 
is the duration of the period of snow cover. A generalised map 
of the average duration on low ground is shown in Fig. 8. Over 
most of the area this is based on actual observations of snow 
cover; for Canada such records could not be found and the 
map was completed by using records of the average duration 
of the period with mean temperature below 32 F. and finding 



44 CLIMATE IN EVERYDAY LIFE 

the relation between the latter figure and the duration of snow 
cover in similar latitudes of northern Europe and Siberia. The 
average duration is less than a month over the western coast of 
North America south of about 55 N., over the United States 
south of about 38°, and over western and southern Europe. 
It increases rapidly northward and eastward and exceeds three 
months over most of Canada, the interior of Norway, Sweden 
north of Stockholm, Finland, the U.S.S.R. and the mountain 
regions of central Asia. 

The scale of the map is too small to show the variations over 
the British Isles. Generally speaking the duration is less than 
5 days a winter over the low ground near the western and 
southern coasts, 5-10 days over most of England and southern 
Scotland, 10-15 over moderately high ground (up to about 
500 feet) in southern and central England and 15-20 over 
similar ground in north-east England, increasing rapidly at 
higher levels. From data given by G. Manley the duration in 
Scotland increases from about 10 days at sea-level to 82 at a 
height of 1,500 feet, an increase of 1 day for each 22 feet of 
height. From 1,500 feet to the summit of Ben Nevis (4,406 
feet, duration 230 days) the increase is at the rate of 1 day 
for each 20 feet of height. In the Alps the duration of snow 
cover increases at the rate of about 1 day for every 33 feet 
of height. In the Pennines the increase is from about 15 at 
350 feet to 76 at 1,840 feet, or 1 day for 24 feet of height. 

Manley also gives the following data about the snow cover 
on some main roads: — 



Table 4. — Snow on main roads, Britain. 

London-Edinburgh (A. 68) . Carter Bar, i ,300 feet . 
London-Glasgow (A.6). Snap, above 1,300 feet 
Stainmore (A.66). 1,400 feet .... 

Manchester-Sheffield (A.57). "Snake" above 1,500 feet 
Perth-Braemar (A. 93). Cairnwell, above 2,000 feet . 
Perth-Inverness (A.9). Drumochter, 1,500 feet 
Glencoe Road (A.82). Rannoch Moor, 1,100 feet 

The depth of the snow cover is so variable from year to year 
that in Britain at least average figures have no meaning. 
R. G. Stone (1940, 1944) gives figures for New England, from 
which the following table is a very brief excerpt. 

Average consecutive number of weeks with 2 inches or 





53 days 




40 „ 




50 „ 




45 >i 




110-120 ,, 




75 » 




55 » 



THE ECONOMICS OF CLIMATE 



45 



20° 40* 60 9 80' IOO° »2Q° I4Q" ICO" 




O* 20° 40° 60" 80° IOO' 120° 140" 160° 



180* I60* 140* 120" lOO' SO* 6O 8 40° 20° 




ISO* \&Q' 140° \ZQ° \QQ' 8Q" 6Q a 4Q° 2Q° 

Fig. 8. — Number of months with snow-cover on low ground. 



46 



CLIMATE IN EVERYDAY LIFE 



more, 5 inches or more, etc., of snow on the ground in New 
England. 

Table 5. — Depth of snow, New England. 



Low ground 

Moderate height (100-200 

feet) .... 

High ground (above 800 feet) 



2 inches 


5 inches 


10 inches 


or more 


or more 


or more 


5 


0-5 


0-4 


15 


10 


5 


20 


15 


10 



15 inches 
or more 



under 5 
5-10 



The average depth in inches of snow at three places on various 
dates is: — 







Oct. 


Nov. 


Dec. 


Jan. 


Feb. 


March 


April 




Height 
feet 














































31 


15 


30 


15 


3i 


15 


3i 


15 


28 


15 


3i 


15 


30 


New York N.Y. 






























(Battery Park) . 


5 


— 


T. 


0-3 


o-8 


07 


1-0 


2-0 


2'0 


1-0 


o-5 


— 


— 


— 


Ithaca N.Y. 


836 


— 


T. 


0-2 


0-3 


o-3 


2-0 


3-0 


4-0 


3-0 


2-0 


I-O 


0-2 


— 


Bethlehem, 






























New Haven 


1440 


~ 


2-0 


3-0 


5-o 


7-0 


9-0 


I2-0 


15-0 


15-0 


13-0 


5-o 


o-8 


~ 



It is necessary to distinguish between two types of snow cover. 
In countries with a long, cold winter, such as eastern Europe, 
northern Asia and the northern interior of North America, a 
snow cover forms early in the winter and persists throughout 
the season. The duration and depth vary considerably from 
year to year, and in the border regions there may be occasional 
thaws, but the general course of events recurs every year. The 
first snow soon becomes compacted and every further fall adds 
to it, to be compacted in turn. The falls are rarely heavy, the 
bulk being made up of frequent light falls. Blizzards occur, 
but these are due more to the raising of the surface snow by 
strong winds than to heavy fresh snow. This type of snow cover 
causes the minimum of interference to traffic; since it is ex- 
pected, provision is made for it such as the substitution of 
sleighs for wheeled traffic. Winter is often the favourite season 
for travel, the frozen snow-covered rivers making excellent 
highways. Where the snow is heavy enough to block railways, 
as in the mountain region of Canada, protection is provided 



THE ECONOMICS OF CLIMATE 47 

in the form of snow sheds. The worst period comes in spring, 
when the snow melts and is replaced by slush, mud and often 
by floods; conditions are especially bad where the ice breaks 
up in the upper reaches of rivers flowing north, while the lower 
reaches are still frozen. This type of snow cover is roughly 
limited by the area within which at least one month has a mean 
temperature below 32 F., shown by the broken line in Fig. 8. 

The second type of snow cover is intermittent. It occurs in 
cyclonic regions where winter temperature oscillates about 
32 ° F. Falls of snow are much less frequent than in the colder 
regions, but when they occur they are often heavy, and are 
accompanied by strong winds which cause drifts. Later there 
is a thaw, which may be accompanied by heavy rain and bring 
severe flooding. This type of snow cover is very irregular in its 
occurrence; whole winters may pass without snow settling on 
the lower ground, while in others there may be repeated falls 
and snow may persist for many weeks. This type of cover 
causes great inconvenience; the deep snow does not have time 
to become compacted and the drifts block roads and railways, 
isolating country districts completely. In cities the roads are 
cleared by snow-ploughs. These snowstorms are best known in 
the north-eastern States of the U.S.A., especially in New York, 
but they also occur in Britain. 

"Glazed frosts" or ice-storms occur occasionally in North 
America and Europe. They take two forms. In the first rain 
falls on frozen roads and similar surfaces, or on compacted 
snow, and freezes. The ice-cover does not often reach any great 
thickness or persist long, and this form is not a serious hindrance 
to traffic. In the second type the raindrops consist of water 
which is still liquid but below its freezing point (supercooled 
water). Each drop freezes immediately it strikes any solid 
object. Roads are covered by a sheet of hard, smooth ice which 
gives no grip, so that wheeled traffic is impossible and even 
walking very difficult. Telephone and telegraph wires become 
thickly coated with ice, which is so heavy in bad storms that 
the wires break under the strain, completing the interruption of 
communications. A severe glazed frost at the end of January 
1940 paralysed large areas of southern England for several days, 
but occurrences on this scale are very rare, probably not oftener 
than once a century, and costly special precautions against 
them are not justified. 



48 CLIMATE IN EVERYDAY LIFE 

Glazed frosts are probably more frequent in the United 
States, where they are known as "ice-storms." They are 
especially frequent in the middle latitudes of the eastern United 
States; one of the worst examples occurred on 26th to 29th 
November 1921, in which wires were coated with cylinders of 
ice 3 inches in diameter. Telephone and telegraph wires, and 
even the posts carrying them, were brought down, the electric 
supply was interrupted and all forms of communication ceased. 
Orchards and even forest trees were destroyed, the total damage 
running into millions of dollars. 

Frost. — Frost is not a direct hindrance to transport, but it has 
a cumulative effect on road surfaces. Ice crystals and layers of 
ice form beneath the surface and break it up. Damage from 
this cause is greater in gravel than in tarred or asphalted roads, 
which hinder the penetration of both water and cold. Wet 
ground freezes to a greater depth than dry ground, and the 
tar or asphalt layer has itself some insulating property. It is 
estimated that the damage caused by frost to the streets of Ohio 
by the severe winter of 1935-6 amounted to 3,000,000 dollars. 
It may be remarked that a snow cover is a good protection 
against damage by frost. Fresh snow, owing to the large amount 
of air which it contains, is the most effective, but even old com- 
pacted snow is a good insulator. 

Frozen earth is also more elastic than unfrozen ground, and 
transmits the vibrations from heavy traffic to neighbouring 
buildings. For this reason it is a good plan to maintain a bed of 
well-worked loose soil, such as a garden, between buildings and 
neighbouring roads carrying heavy traffic. 

Interruption of Navigation by Sea Ice. — The most serious effect 
of winter cold on transport is the blocking of rivers, harbours 
and coasts by floating ice. The areas chiefly affected are: (1) 
South-eastern Canada and the St. Lawrence Estuary; (2) 
Newfoundland, Labrador and Hudson Bay; (3) the White Sea 
and the Arctic coast of the U.S.S.R. ; (4) the coasts of the 
Baltic; (5) the coasts and river mouths of the Black Sea; (6) 
the Sea of Okhotsk, and neighbouring coasts. 

( 1 ) The rivers of eastern North America and the landlocked 
harbours north of about latitude 45 ° freeze in winter, and in 
the smaller ports navigation is interrupted for two or three 
months. There is some freezing of rivers down to latitude 37 
or 38 . The more important ports are generally kept open by 



THE ECONOMICS OF CLIMATE 49 

ice-breakers or by the normal traffic. Table 6 gives the average 
dates of closing and opening of a number of ports ; fuller details 
are given in the Ice Atlas of the Northern Hemisphere, published by 
the Hydrographic Office, Washington, price eight dollars. In 
the ports which are blocked each winter the regular steamer 
services are usually withdrawn two or three weeks before the 
average date of freezing, and resumed a few weeks after the 
thaw. Lake Erie is heavily iced in January and February. 

The narrow entrance to the St. Lawrence Estuary through 
Cabot Strait, between Cape Breton Island and Cape Ray in 
Newfoundland, is a great obstacle to shipping in winter. From 
about mid-January to late April drift ice makes the Strait 
impassable to ships not specially built to encounter ice. Towards 
the end of March the winter ice of the St. Lawrence River and 
the Great Lakes begins to break up, and from mid-April to 
mid-May there is a great rush of drift ice down the river. In 
many years Cabot Strait is jammed by ice, sometimes for as 
long as three weeks, between mid-April and mid-May, and 
completely closed to shipping. This ice-jam is known as the 
"Bridge"; it has caused a number of wrecks off the coast of 
Newfoundland . 

(2) Newfoundland and Labrador. The river mouths and 
harbours of Newfoundland (except the south coast) and 
Labrador freeze for a time in winter; the average dates of 
closing and opening of ports are given in Table 6. A more 
serious obstacle on the east coast is the drift ice. During the 
winter Baffin Bay and all the western side of Davis Strait are 
filled with unnavigable ice until the end of June; this ice con- 
sists of floes in which icebergs are frozen. In the early spring 
this mass of ice drifts slowly south-eastwards in the Labrador 
Current, entering the Atlantic off Newfoundland. Here the 
ice-floes disintegrate in the warmer water, but the icebergs 
drift about off Newfoundland and sometimes penetrate far 
into the Atlantic, where they constitute a grave danger to 
shipping. On 14th April 191 2 the s.s. Titanic struck a berg in 
lat. 4i°46 / N., long. 50°i4' W., and sank with great loss of life. 
The amount and date of arrival of the ice vary greatly from 
year to year, but bergs frequently appear in numbers off die 
Newfoundland Banks in the second half of March. 

Hudson Strait and Hudson Bay. Ice begins to form on the 
north shore of Hudson Strait and the shores of Hudson Bay in 



50 CLIMATE IN EVERYDAY LIFE 

October. Hudson Strait generally becomes unnavigable except 
by powerful ice-breakers in November, and in the same month 
fast ice has spread to all shores of Hudson Bay, though it is still 
thin in the southern half. The whole coast of Hudson Bay is 
blocked in December. The ice begins to break up in May and 
Hudson Strait is generally navigable by powerfully built vessels 
in June. In July all ice has practically disappeared. The 
Churchill and Nelson Rivers are closed to navigation from 
early in November to mid-May, and the Albany River from mid- 
November to early May. 

(3) The White Sea and Arctic coast of Russia. Ice may form 
in the Gulfs of Kandalaksha, Onega and Archangel as early as 
October and regularly in November, but the northern part of 
the White Sea is not generally beset until December. The ice 
spreads westward along the coast until March, by which time 
it has generally reached Sviaitoi Nos; after this it retreats. 
The whole of the White Sea is covered by drift ice from January 
to April inclusive, but the central part is usually navigable 
by ordinary vessels in January and by heavily built ships from 
February to April. The ice becomes very loose in May and 
disappears in June. The Murman Coast is generally ice-free 
throughout the winter. The occurrences of ice are very irregu- 
lar; in Table 6 the first and last dates over a period of about 
seventeen years are given. 

(4) The Baltic. The western Baltic is not frozen regularly. 
In very severe winters ice forms in the estuaries south-west of 
Denmark, blocking Bremen and Hamburg, and somewhat more 
frequently farther east, but even ports as far east as Konigsberg 
are affected only one year in five. Entrance to the Baltic 
through the narrow straits of Denmark is only blocked in 
severe winters (not more than twice in ten years). The ice 
reaches its maximum thickness in February, and except in the 
severest winters the straits are clear by the end of March. 

The coasts of the Gulf of Bothnia generally freeze in 
December or January. The southern half is freed about 
the end of April, but north of 64 N. the ice persists well into 
May. The ports from Umea round to Vaasa are closed every 
year, Hernosand one year in three and the Swedish coast 
between Hernosand and 6o° N. only occasionally. Access to 
Stockholm is rarely hindered by ice, but ice sometimes blocks 
the narrow strait between Oland and Kalmar. On the 



THE ECONOMICS OF CLIMATE 5 1 

Finnish side Rauma is occasionally beset, but Abo (Turku) is 
not affected. 

In the Gulf of Finland in average years ice appears on the 
north coast at the end of November, and by nth December 
there is an even chance that the Bay of Kronstadt will be 
partly frozen. The limit of freezing in average years spreads 
gradually along the south coast during January and February, 
reaching its most westerly point, between 2i° and 22° E., early 
in March. By this time the whole of the Gulf of Riga and the 
sea surrounding the islands of Hiiumea and Saaremaa is 
covered by unnavigable ice. From the beginning of April the 
ice breaks up rapidly, and the whole Gulf is generally free by 
the beginning of May. Viipuri and Kronstadt are closed every 
year, Helsinki about one year in two or three. Off south-west 
Finland navigation is not interrupted. 

(5) Black Sea and Sea of Azov. Only the northern part of 
the Black Sea is affected by ice. The Lower Danube is frozen 
regularly, especially the delta round Galatz and Braila; the 
ice breaks up very rapidly at the end of February. In the north- 
west Black Sea all lagoons and river mouths are frozen regu- 
larly for a long period. On the south coast of the Crimea icing 
is slight, and the Strait of Kerch is not usually much affected 
until the end of January, when a wide belt of pack ice forms at 
the northern entrance. The Sea of Azov is the worst sufferer; 
ice-breakers keep the ports open during the first and last weeks 
of the ice season, but in the main cold period during February 
the ice is too thick. 

(6) East Coast of Asia and Sea of Okhotsk. In the Yellow 
Sea ice occurs only in the bays on the west coast of Korea north 
of 37 N. in the second half of January and beginning of 
February; ice practically never appears in Korea Strait, and 
the Japan Sea proper is ice-free throughout the winter except 
for a narrow strip of drift ice from the Sea of Okhotsk lying off 
the Russian coast from Vladivostok northwards in January, 
February and the first half of March. Navigation is always 
possible south of 48 N. and Vladivostok is kept open by ice- 
breakers throughout the winter. On the east side of the Japan 
Sea there is little ice. The Gulf of Tartary, off Nikolaevsk, is 
blocked by compact ice by mid-December, and remains so 
until May or even early July. 

In the Sea of Okhotsk ice formation begins at the end of 



52 CLIMATE IN EVERYDAY LIFE 

October or early November on the north shore; it persists until 
June, when in places it is only navigable by heavily built vessels, 
and some ice remains in July. Near the southern Kuriles drift- 
ice is met in March to May, but there is only scattered ice off 
the west coast of Kamschatka. Off eastern Kamschatka there 
is drift-ice and some fixed ice ; access to Petropavlovsk is main- 
tained by icebreaker. Farther north the Gulf of Anadyr is 
blocked from late November to early May, and Bering Strait 
from mid-October until late May. 

The "Debacle" — The spring break-up of the ice in the rivers 
of Canada, Russia and Siberia is an imposing affair, which is 
termed the "Debacle." In southern Canada and southern 
Russia it begins about the middle of March, but is progres- 
sively later in higher latitudes, not coming until May or even 
June in the extreme north. It lasts from a fortnight to six weeks, 
during which time the drifting ice masses often form jams, 
blocking the rivers and causing floods. In the northward flowing 
rivers of Siberia, in which the upper courses melt first and flow 
over the still frozen lower reaches, the whole country becomes 
impassable; Irkutsk, for example, is isolated for a month in 
May. 



THE ECONOMICS OF CLIMATE 



53 



Table 6. — Dates of Closing and Opening of Ports, etc. 



Place 



Nova Scotia — 
Shelburne 
Yarmouth 
Halifax Hbr. 

St. John Hbr. 
Gut of Canso 
Pictou Hbr. 
Amherst Hbr. 
Port Hood 
Sydney 

New Brunswick — 
Shediac . 
Chatham . 
Miramichi Bay . 
Campbellton 

Prince Edward Is. — 
Charlottetown . 
Summer side 

Quebec — 
Father Point 
Gaspe Hbr. 
Seven Is. . 

St. Lawrence Gulf— 
St. Pierre Road- 
stead . 
Cabot Strait 

Newfoundland — 
Port au Basques 



La Poile Hbr. . 
Burgeo Port 
Grand Bank Hbr. 
Burin Hbr. 
Trepassey Hbr. . 
Hearts Content 

Hbr. . 
Harbour Grace 
Cape Race 
Conception Bay 
St. John's Hbr. . 



Lai. 

N. 


Long. 
W. 


Date. 


Of 


Closing 


Opening 


43 47 

43 55 

44 35 


65 19 
65 45 
63 33 


— 


— 


45 16 
45 30 
45 40 

45 49 

46 2 
46 7 


66 4 

61 8 

62 38 
64 13 
61 31 
60 18 


Jan. 1 
Jan. 1 
Dec. 15 
Jan. 30 
Jan. 10 


April 30 
April 10 
April 10 
May 1 
April 1 


46 12 

47 1 

47 4 

48 


64 27 

65 26 

65 15 

66 34 


Dec 1 
Dec. 8 
Dec. 12 
Dec. 1 


April 1 
April 18 
April 13 

May 1 


46 14 
46 23 


63 8 
63 47 


Dec. 21 
Dec. 20 


April 20 
April 10 


48 32 
48 52 
50 13 


68 28 

64 30 

66 24 


Jan- 15 
Dec. 15 
Nov. 1 


March 1 5 
May 10 
April 15 


46 45 

47 30 


56 12 
60 


Jan. 15 


April 30 


47 33 


59 10 


— 


— 


47 45 
47 35 
47 4 
47 * 
46 40 


58 18 
57 40 
55 46 
55 H 
53 21 


— 


— 


47 52 
47 39 

46 38 

47 50 
47 33 


53 22 
53 16 
53 3 
52 50 
52 40 


Feb. 10 

Feb. 1 
Feb. 10 


March 31 

May 1 
April 5 



Remarks 



Seldom interrupted 
Not hindered. 
Frozen three times 

in ninety years. 
Not hindered. 



Seldom interrupted. 

Ice "Bridge" about 

April 15-May 15. 

Harbour sometimes 
blocked by drift 
ice Feb.-March. 

6 springs in 50. 

Seldom interrupted. 

Twice in 30 years. 

Seldom interrupted. 

Seldom interrupted. 



Not hindered. 



Rarely closed for a 
week in an aver- 
age season. 



54 



CLIMATE IN EVERYDAY LIFE 









Dates of 




Place 


Lat. 

N. 


Long. 
W. 






Remarks 












Closing 


Opening 




Newj ' dland — ctd. — 












Trinity Hbr. 


46 40 


53 21 


— 


— 


Seldom interrupted. 


Catalina Hbr. . 


48 31 


53 4 


Jan. 1 


April 15 




Bonavista Bay . 


48 55 


53 10 


— 


— 


Generally passable. 


Cape Freels 


49 16 


53 27 


Jan. 15 


May 10 




Gander Bay 


49 20 


54 25 


Jan. 1 


May 1 




Fortune Hbr. . 


49 32 


55 !5 


Dec. 15 


May 7 




St. Anthony Hbr. 


51 21 


55 30 


Nov. 30 


May 15 




Belle Isle Str. . 


5i 35 


5 6 15 


Dec. 20 


April 1 


First steamers be- 
tween June 7 and 
July 25, last Nov. 
1 1-26. 


Bay of Islands . 


48 57 


57 55 


Jan. 5 


April 30 




St. George Bay . 


48 50 


58 


Jan. 15 


April 18 




Labrador — 












Alexis R. . 


52 30 


56 


Dec. 25 


May 10 




Lake Melville . 


53 4o 


59 40 


Dec. 25 


April 15 


Heavy pack ice off 
approaches pre- 
vents navigation 
until well into 
June. 


Cartwright Hbr. 


53 42 


56 58 


Nov. 25 


May 15 




Hamilton Inlet . 


54 


59 


Nov. 30 


June 15 


Mail boat operates 
July i-Oct 15. 






Long. 


Mean 


f ates of 








E. 


Freeze 


Thaw 




Arctic Coast 












Finland — 












Kemi 


65 41 


24 42 


Nov. 6 


May 25 


Navigation closed. 
Nov. 25-May 22. 


Russia — 












Murmansk 


69 


35 


Very ir 


regular 


Generally ice-free. 


Archangel 


64 33 


39 40 


Oct 23- 

Nov. 25 


May 6- 
30 




Mezen 


65 50 


44 23 


Oct 20- 

Nov. 12 


May 5- 
22 




Baltic 












Sweden — 












Hernosand 


62 37 


17 58 


Jan. 25 


April 13 


Feb. 7-April 16. 


Umea 


63 45 


20 20 


Nov. 12 


May 9 


Dec. 23-May 1. 


Lulea 


6 5 35 


22 10 


Dec. 25 


May 14 




Finland — 












Tornio 


65 48 


24 15 


Nov. 30 


May 24 




Oulu (Uleaborg) 


65 3 


25 35 


Dec. 10 


May 15 




Vaasa 


63 38 


21 42 


Jan. 15 


April 12 




Abo (Turku) . 


60 26 


22 18 


Dec. 5 


April 17 


Navigation not in- 
terrupted. 



THE ECONOMICS OF CLIMATE 



55 









Mean dates of 




Place 


Lat. 


Long. 






Remarks 


N. 


E. 


Freeze 


Thaw 


Finland — contd. — 












Helsinki 












(Helsingfors) . 


60 9 


24 57 


Dec. 1 1 


April 22 


Feb. 16-April 5. 


Hango 


59 49 


23 


Jan. 7 


April 7 


Navigation not in- 
terrupted. 


Viborg (Viipuri) 


60 55 


28 30 


Nov. 26 


April 28 


Jan. 1 -April 20. 


Russia — 












Kronstadt 


59 59 


24 49 


Dec. 5 


April 29 


Jan. 3 1 -April 22. 


Estonia 












Revel (Tallinn) 


59 24 


24 25 


Feb. 2 


April 3 


Not interrupted. 


Latvia — 












Riga Hbr. 


56 57 


23 30 


Dec. 26 


April 6 


March 8- April 2. 


Windau Hbr. . 


57 24 


21 32 


Jan. 12 


Feb. 27 


Not interrupted. 


Libau Hbr. 


56 30 


21 


Jan. 8 


March 1 


Not interrupted. 


Black Sea 












Roumania — 












Braila 
Galatz 


45 16 
45 28 


27 58 

28 4 


} Jan. 3 


March 1 


Jan. 8-Feb. 27. 


Sulina 


45 10 


29 40 


Jan. 7 


Feb. 17 


Kept open by 
breaker except in 
severe winters. 


Russia — • 












Odessa 


46 30 


30 40 


Jan. 10 


March 15 


Ditto 


Nikolayev (Bug) 


46 58 


32 


Dec. 10 


April 1 


Dec. 25-March 20. 
Kept open by ice- 
breaker in mild 
winters. 


Kherson 












(Dnieper) 


46 39 


32 38 


Dec. 14 


March 25 


Dec. 24-March 18. 
Kept open by ice- 
breaker in mild 
winters. 


Kerch 


45 21 


36 28 


Dec. 28 


March 15 


Kept open except in 
severe winters. 


Rostov-on-Don . 


47 15 


39 40 


Dec. 6 


March 27 


Dec. 14-March 27. 



CHAPTER II 

THE SITING AND DESIGN OF HOUSES AND 
FACTORIES IN RELATION TO CLIMATE 

ANY building must be designed to fit its surroundings, and 
AA of these climate is the most important. If the house or 
*■ ^ factory is built to withstand rigours which rarely or never 
occur, it will be too expensive to build. If, on the other hand, 
it is too flimsy it will be expensive to heat and maintain. The 
factors to be considered are : — 

( i ) Gain or loss of heat through walls, windows and roof. 

(2) Lighting. 

(3) Purity of air. 

(4) Local situation in regard to (a) frost and fog; (b) rain; 

(c) snow. 

(5) Wind (wind pressure on buildings, dissipation oPsmoke). 

(6) Rain and wind together (leaking of walls) J 

A. Geddes (1946) recommends that the distribution of the 
climatic factors should be sketched in on contour maps on a 
scale of 1 inch or 6 inches to the mile, but he points out that 
not only the existing climate is to be considered, but what it 
will be after building is completed. 

The constructional details are, of course, a matter for the 
architect; the following pages give a summary of the necessary 
climatic basis, as far as it is available. 



GAIN OR LOSS OF HEAT THROUGH WALLS, WINDOWS AND ROOF 

The heat balance of a building is very complicated. A 
building is heated by the absorption of solar radiation and, in 
hot weather, by conduction from the air. It is cooled by radia- 
tion from the walls and roof, by evaporation from wet surfaces, 
and, in cold and especially windy weather, by conduction to 
the air. In a hot climate the problem is to resist the absorption 
of external heat, especially of direct radiation from the sun; 
in a cold climate the problem is to conserve the heat, produced 
internally, against conduction to the external air. 

56 



SITING AND DESIGN OF HOUSES AND FACTORIES 57 

Solar Radiation. — The amount of heat received from the sun 
depends on the elevation of the sun above the horizon, the 
amount of cloud, and the purity of the air. At the outer limit 
of the earth's atmosphere the sun's rays have a heating power 
of nearly two calories per sq. cm. per minute, equivalent 
to one B.T.U. on 20 square inches. Part of this heat is 
absorbed by water vapour and other gases in the atmosphere, 
part is reflected from the upper surfaces of clouds, and part is 
scattered by the molecules of air and dust particles in the 
atmosphere. 

The visible part of the sun's rays accounts for the greater 
part of the heating effect, but there are also rays invisible to the 
human eye. At one end of the scale these include the ultra- 
violet radiation, which has powerful chemical effects. These 
rays cause chemical changes in the skin which are visible as 
sunburn, but they also have health-giving properties. It is the 
shutting-off of the ultra-violet radiation by smoke and fog 
which is mainly responsible for the pallid appearance of city 
dwellers, and for the tendency to rickets among city children. 
The direct heating power of these rays is small. At the other 
end of the scale the infra-red (sometimes called ultra-red) rays 
have considerable heating power; these are not much affected 
by smoke, but are powerfully absorbed by water vapour. 

The upper surfaces of thick clouds reflect about three- 
quarters of the sun's radiation back to space. This means a 
great loss of heating power, which is accentuated because in 
cloudy regions the amount of water vapour in the air is also 
relatively great. Hence the highest day temperatures, exceeding 
1 30 F. in the shade, are found in dry climates in tropical and 
sub-tropical regions where there is little or no cloud. In moist, 
cloudy, equatorial regions the air temperature rarely rises 
above ioo° F. 

On the other hand, both water vapour and cloud have a 
great effect in conserving heat at night. The earth gives back 
at night and in winter the heat it has absorbed from the sun 
by day and in summer. This heat is given up as infra-red 
radiation, which is rapidly absorbed by water, whether a> 
vapour or cloud droplets. In a moist climate the heat cannot 
readily escape and the nights are warm and muggy. In a dry 
climate the heat gained during the day is rapidly lost after 
sunset, and the nights are cold. 1 For this reason buildings in hot 



58 CLIMATE IN EVERYDAY LIFE 

dry climates should have thick walls, shaded by wide verandas, 
and thick roofs, preferably double, to equalise the temperature 
of the interior. Windows should as far as possible be sheltered 
from direct sunshine by verandas and, outside the tropics, by 
being placed mainly on the north side (in the northern hemi- 
sphere). Apart from this, ample opening space of windows 
facilitates ventilation during the cooler times of day. The type 
of building favoured by local residents should be studied, as this 
has been developed as a result of long experience. In hilly 
countries with hot summers it may be possible to select a 
house-site to take advantage of the cool breeze which blows 
down the hillsides at night, and to make the best use of it by 
building summer sleeping porches on the upslope side of the 
house. 

The neighbourhood of small lakes is an unfavourable situation 
in places with hot summers. A small or shallow body of water 
reaches a high temperature, and active evaporation makes the 
lowest layers of air very humid, especially in calm, sunny 
weather. On the other hand, lakes large and deep enough to 
absorb the sun's heat without warming up appreciably have a 
cooling and moderating effect on the air. 

The effect of position in a building was shown by A. J. ter 
Linden (1938), who obtained autographic records of the indoor 
temperature and relative humidity of a number of rooms in 
Delft, Holland, during the rather cool summer of 1937, with 
the outdoor temperature for comparison. In a room on the 
ground floor of a large building with windows facing east and 
north, the temperature was very steady at about the average 
of the outdoor temperature ; relative humidity was also steady, 
between 60 and 70 per cent. At the other extreme, in a room 
with large windows north and south, immediately under a 
flat roof, a marked "greenhouse" effect was found; the room 
temperature ranged from 66° to 76 F., while the outdoor 
temperature was between 46 and 64 F. The indoor maxima 
and minima lagged two or three hours behind those of the 
exterior. Relative humidity remained around 50 per cent. 
A room immediately below a flat roof, but with windows only 
to the east, showed an intermediate effect. CTrie type of roof is 
important; ter Linden points out that in nbt weather gabled 
roofs with a good airspace below give much more comfortable 
conditions in the rooms on the top floor than do flat roofs. If 



SITING AND DESIGN OF HOUSES AND FACTORIES 59 

the latter are insufficiently insulated the top story is likely to 
be unhealthily hot and dry) 

No observations seem to exist of temperatures inside buildings 
with walls of different thicknesses, but an idea of the effect may 
be obtained from soil temperatures at different depths. On 
Salisbury Plain N. K. Johnson and E. L. Davies (1927) found 
that the logarithm of the daily range of temperature is pro- 
portional to the depth below one inch. At 5 inches the range 
was reduced to one-tenth, and at 10 inches to one-hundredth, 
of that at one inch. The character of the soil made little differ- 
ence. At Cairo, on a hot day on which the surface of sandy 
soil had a daily range of 57 F., the range was only 6-5° F. at 
a depth of 8 inches. From this it is seen that the high tem- 
peratures of earthy or stony materials exposed to the sun are 
very^superficial. 

{The part played by the colour of the walls in modifying the 
effect of the sun's heat may be judged from some experiments 
at Poona in India. Black soil in May reached a temperature 
of 64-6° G. (148 F.), brown soil of 58 G. (136 F.), and soil 
covered with white powder only 48 G. (118 F.). Throughout 
the tropics the surfaces of dark-coloured bricks, tarred roads, 
etc., exposed to the sun reach temperatures of 130-140 F., and 
may exceed 160 in hot, dry regions. It must be remarked, 
however, that while white walls and roads are most efficient in 
reflecting the sun's heat, they also reflect the glare and are very 
exhausting to the eyes. For that reason the tendency is to replace 
white by some light colour. Most roofing materials in use 
absorb about 90 per cent, of the sun's radiation; galvanised 
iron absorbs roughly two-thirds. The effect of colour is discussed 
mpjqe fully in Chapter IX. 

t/The effect of the elevation of the sun on the intensity of solar 
radiation is very great. In passing through a unit thickness of 
air the rays lose a certain proportion of their power. The unit 
adopted is the thickness of air traversed by the sun when it is 
vertically overhead; this is termed "air mass 1." In clear, dry 
weather with a vertical sun about 85 per cent, of the solar 
energy reaches low ground; in damp, humid weather such as is 
generally found near the Equator the ratio is about 75 per cent. 
In large to^vns with a smoky atmosphere the loss of energy is 
much greater) 

The thickness of air traversed when the sun is at different 



6o 



CLIMATE IN EVERYDAY LIFE 



altitudes, and the maximum solar radiation reaching the ground 
in dry and humid air (transmission coefficients 85 and 75 per 
cent.) are shown in Table 7. The figures in lines (3) and (4) 
refer to a surface at right angles to the sun's rays; lines (5) and 
(6) show the radiation reaching a horizontal surface. The figures 
are in calories /cm. 2 /min., assuming that the value at the limit 
of the atmosphere is 1 -93. The lowest two lines show the direct 
solar radiation falling on a vertical wall facing south (in the 
northern hemisphere). 



Table 7. — Height of sun, air mass and heat received. 



( 1 ) Height of sun 

(2) Air mass 
Radiation at sea 


90° 

I'O 


70° 
1 -064 


50° 
1-305 


30° 
J-995 


20° 

2-9 


15° 
3-81 


IO° 

5-6 


5° 
10-4 


level. 


















Normal to rays. 

(3) Dr Y air 

(4) Humid air 
Horizontal surface. 


1 64 
*'45 


1-62 
1-41 


1 -56 
i'33 


1-4 

1-09 


I-2I 

0-84 


1-04 
0-65 


0-78 
o-39 


0-36 
o-i 


(5) Dry air 

(6) Humid air 
Vertical S. wall 


1-64 
i-45 


1-52 
i-33 


1-2 
1-02 


0-7 
o-54 


0-41 

0-29 


0-27 
0-17 


0-13 
0-07 


0-03 
o-oi 


(7) Dry air 

(8) Humid air . 


o-o 
o-o 


0-56 
0-48 


1-0 
0-8 5 


I-2I 

0-94 


1-13 
0-79 


1 -oi 
0-62 


077 
038 


o-35 
o-i 



At noon in March and September the elevation of the sun is 
equal to 90 minus the latitude. At midsummer outside the 
tropics it is 1 13 minus the latitude, and in midwinter 67 minus 
the latitude. North of 67 the sun does not appear at all in 
midwinter. The altitude of the sun at noon on any day is equal 
to 90 minus the positive difference between the latitude and 
the sun's declination. 

The effect of the greater absorption by the air in higher 
latitudes, due to the greater thickness of air traversed, is mag- 
nified by the fact that, owing to the greater inclination of the 
sun's rays on level surfaces the same amount of energy is spread 
out over a greater area. The decrease in the amount of heat 
from the sun received on the ground and on flat roofs in higher 
latitudes is partly compensated by the gain on vertical walls 
facing south, but this gain is small. Buildings are warmed not 
only by direct sunshine, but also by heat reflected or radiated 
from the ground, and this depends on the angle of the sun's rays. 

The sun's declination at noon on the middle day of each 



SITING AND DESIGN OF HOUSES AND FACTORIES 



6l 



month, and the noon altitude of the sun in 50 N., are approxi- 
mately as follows : — 



Table 8. — Sun's declination and altitude. 



Solar Declination — 



Jan. 15 


Feb. 14 


March 15 


April 1 5 


May 15 


June 15 


July 15 


Aug. 15 


Sept. 15 


Oct. 15 


Nov. 15 


Dec. 15 


-21 19' 


-13° 23' 


-2° 34' 


+ 9° 22' 


+ i8° 3 6' 


+ 23 16' 


+ 21° 42' 


+ 14° 24' 


+ 3° 27' 


-8° 6' 


-18 I2'|— 23 13' 


Noon Altitude in 50 N. 


Jan. 15 


Feb. 14 


March 15 


April 15 


May 15 


June 15 


July 15 


Aug. 15 


Sept. 15 


Oct. 15 


Nov. 15 


Dec. 15 


19 


27° 


38° 


49° 


59° 


63° 


62° 


54° 


43° 


32° 


22° 


17° 



The effect of solar heating is modified by the elevation of the 
ground and its slope. Over high ground the amount of air is 
less than over low ground, and its power of absorption is 
correspondingly less. That accounts for the strong sunshine of 
mountain resorts, which goes a long way towards counter- 
balancing the low air temperature. At a height of 3,000 feet 
the air mass is reduced by about one-tenth, so that at 3,000 
feet in latitude 50 the strength of the sun's rays is equivalent 
to that at sea-level in latitude 45 . In humid or smoky places 
the effect is greater than this because much of the moisture 
and most of the smoke is confined to the lowest tenth of 
the air. 

( The effect of a southerly slope (in the northern hemisphere) 
is "In some ways similar to transfer to a lower latitude. The 
radiation received on flat roofs and vertical walls is of course 
unchanged, but the radiation received on the ground is in- 
creased. Ground sloping south becomes warmer than level 
ground, and part of this extra heat is given up to the buildings. 
Also the area shaded by the buildings is smaller. On ground 
sloping north the effect is reversed. In middle northern lati- 
tudes the effect of a south slope of one in ten on clear days is 
to increase the heating effect on the ground by about 15 per 
cent. ; a slope of one in ten to the nortii decreases the heating 
effect by the same amount. On cloudy days there is little 
difference between different slopes. On a southerly slope there 
is some loss of radiation in summer before 6 a.m. and after 
6 p.m., but this is relatively unimportant. The effect of an 



62 



CLIMATE IN EVERYDAY LIFE 



easterly slope is to increase the warming effect in the morning 
and decrease that in the afternoon; the "radiation day" both 
begins and ends earlier. Similarly, the effect of a westerly slope 
is to make the radiation day begin and end later. A northerly 
slope is altogether unfavourable, except in the early morning 
and Jate evening. 

(The total radiation received on vertical walls depends on the 
season and the direction in which the wall faces. For the year 
as a whole it is greatest on walls facing south and least on walls 
facing north, but in summer, south of about 55 N., the sun is 
so high at noon that the direct heating effect on vertical walls 
is small, and for this reason most heat is received on walls facing 
south-east and south-west. 

J. M. Stagg (1950) gives data for the components of direct 
solar radiation which fall on surfaces facing in different direc- 
tions at Kew Observatory, Richmond, representative of the 
outskirts of London. The following table, based on data for 
all days in the period 1933 to 1946, has been constructed from 
these data (unit, gm.cal./cm. 2 /day). 



Table 9. — Daily averages of direct radiation on differently 
oriented surfaces. 



Month 


Jan. 


Feb. 


Mar. 


Apr. 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Mean 


Normal to sun 
Vertical walls facing 

south 

north 

east or west 
Horizontal surface 


38 

33 

7 
9 


72 

55 

17 
23 


130 

78 

36 
55 


192 

80 

2 

58 

102 


258 

74 

11 

79 

153 


294 

66 

20 

91 

182 


245 

56 

13 

70 

i39 


234 

73 

4 

64 

121 


181 
93 

52 

86 


108 
75 

27 

40 


49 

4i 

10 
13 


36 

31 

7 
7 


153 

63 

4 

43 

77 



One gm.cal. equals 4-18 joules, or 0-00397 B.T.U. Thus 
a radiation of 1 gm.cal. /cm. 2 equals 3-7 B.T.U. per square 
foot. 

Radiation on a wall facing in any other direction, or on a 
slope, is readily calculated by combining the figures from the 
directions on either side. For example, a wall facing SSE. 
makes an angle of 22|° with a wall facing south and an angle 
of 67J with a wall facing east. To find the radiation on such 
a wall, we multiply the radiation on a south wall by the cosine 
of 22J or 0-924, and that on an east wall by the cosine of 
67^° or 0-383. The resulting figures are squared and added and 
the square root taken. Thus, in June the radiation on a wall 



SITING AND DESIGN OF HOUSES AND FACTORIES 63 

facing SSE. is -\/[(66x 0-924) 2 + (91 Xo-383) 2 ] = 7o gm.cal./ 
cm. 2 /day. On the clearest days the radiation is much stronger, 
amounting in June to 1,035 gm.cal./cm. 2 normal to the sun, 642 
on a horizontal surface and 319 on a wall facing east or westy 

In addition to direct solar radiation the earth receives radia- 
tion from the sky. This takes two forms: (1) diffuse short- 
wave radiation, which is simply solar radiation scattered by 
water drops, dust particles and gas molecules during its passage 
through the air; (2) long- wave radiation from the clouds and 
the air itself. Both these types of radiation come from all 
directions, though the diffuse radiation is greatest from the 
neighbourhood of the sun and least from near the horizon. 
The diffuse short-wave radiation on a horizontal surface at 
Kew ranges from about twice the direct solar radiation on a 
horizontal surface in winter to about equal to the latter in 
summer. It is greatest on thinly clouded days and least on 
heavily clouded days. Since a vertical wall receives radiation 
from only half the sky, the diffuse radiation on a wall facing 
in any direction may be taken as about half that falling on a 
horizontal surface, i.e. about equal to the bottom line of Table 9 
from October to March, half the bottom line from April to 
August, and four-fifths of it in September. 

The long-wave radiation from air and clouds is small on 
clear days but becomes more important on cloudy days, when 
it amounts to about 12 gm.cal./cm. 2 /day on a horizontal surface. 
Allowing for radiation from the ground, the total long-wave 
radiation on one side of a vertical wall may be taken as ratiier 
less than this in winter and rather more in summer. 

Measurements of direct and diffuse solar radiation are not 
available for many places, but an approximate calculation of 
the direct solar radiation under clear skies in any latitude can 
be made without much difficulty. The "solar constant" is 
1 -93 gm.cal./cm. 2 /min.; this is the radiation at the limit of die 
earth's atmosphere. The transmission coefficient, or the pro- 
portion of radiation reaching sea-level under a vertical sun, may 
be taken as about 0-85 in dry regions, about o-8 in average 
climates such as prevail in most of Europe and North America, 
and 0-75 in moist equatorial climates and in the rainy season 
in monsoon climates. If we write / for this transmission coeffi- 
cient and a for the air-mass, or length of the path of the sun's 
rays through the atmosphere, while / (in gm.cal./cm. 2 /min.) 



64 CLIMATE IN EVERYDAY LIFE 

is the intensity of the solar radiation on a plane normal to the 
sun's rays at the surface of the earth, then 

log /= 0-28564-0 log t. 

The air-mass a depends on the zenith distance z of the sun, and 
is approximately equal to sec z- z can be found from the 
following expression: — 

cos £=cos P cos <j) cos S-j-shi <j> sin S. 

Here <f> is the latitude, 8 is the sun's declination, and P is the 
"hour-angle," i.e. the number of hours before or after local 
noon, converted to angular measure at the rate of 15 per hour. 

Having found / for, say, each hour, the values can be split 
up into their components on a horizontal plane and on walls 
facing south, east and west, multiplied by sixty to convert to 
radiation per hour, and by any other factors to bring to the 
units required, and added together to give the total radiation 
during the day. This method was tried out for latitude 52 ° N. 
for 15th March, using a transmission coefficient of o-8, and 
gave totals about equal to those found by Stagg on bright days 
in March. 

The elevation and azimuth of the sun at any time can be 
found graphically by the use of a simple diagram. A convenient 
one was published by the U.S. Hydrographic Office on the 
back of the Pilot Chart of the Central American Waters for May 
1 94 1 ; a simplified and reduced version of this is reproduced in 
Fig. 9 to illustrate the principle, but for accurate work the 
original should be consulted. 

To find the height of the sun at any time, first find the 
latitude of the place, and the declination of the sun on the day 
in question. Add these and find the corresponding point G x 
on the upper scale of the triangle. Subtract them and find the 
corresponding point G 2 on the lower scale. Draw straight lines 
from C x to A x and from C 2 to A 2 and mark the point of inter- 
section X of these two lines (the sloping lines on the triangle 
all intersect at A x or A 2 , so that the point of intersection can 
be found by interpolating between these lines). Next find the 
hour-angle of the sun, i.e. the difference in hours between the 
time required and local noon, multiplied by 15. For accurate 
results a correction must be made for the equation of time, or 
the difference between noon by local time and apparent noon 



SITING AND DESIGN OF HOUSES AND FACTORIES 65 

by the sun. Sun time (as shown, for example, by a sundial) 
is behind clock time in January, February and March, the 
difference reaching nearly 15 minutes in February, and again 
by a small amount in July and August. Sun time is "fast" by 
the clock by a small amount in May and again in September to 
December, the difference exceeding 15 minutes from about 

A2 

20 
30 

40 




Fig. 9. — Diagram for height and aximuth of the sun. 

20th October to 17th November. The corrections to clock 
time are to the nearest half-minute on the 15th of each month 
(data for 1 946) : — 

Jan. Feb. Mar. April May June July Aug. Sept. Oct. Nov. Dec. 
— 9i — i4i —9 0+4 o —6 — 4$ +4i +14 +15$ +5 

Having found the hour-angle of the sun, mark this point on 
the line A X A 2 and draw a line from it to the point X. The 



66 CLIMATE IN EVERYDAY LIFE 

intersection of this line with the scale go—h gives the distance 
of the sun below the zenith, from which the height h of the 
sun can be obtained directly. 

To find the azimuth of the sun, having first found h, mark 
the point Y in the triangle representing 0-f A and <f>— h, and 
join this to the point on the h scale given by 90— o\ The inter- 
section of the extension of this line with the go—h scale gives 
the azimuth in degrees from north through east (morning) or 
west (afternoon). 

Example. — Consider a point in lat. 52 ° N. at 9 a.m. local time 
on 14th February. The sun's declination 8 is — 13 , so that 
^-f 8 is 39 and <f>— 8 is 65 . The sun time is nine hours less 
nearly 15 minutes (equation of time), so that the hour-angle 
is I 5X3i=49°' Joining these two points we find that 90— h= 
77°, so that A=i3°. ^+^=65° and <£— A=39° in the triangle 
and 90—8=103° on the h scale. Joining these points and 
extending to the A scale we find that the azimuth is 132°, i.e. 
the sun is 132—90=42° south of east. 

Alternate exposure to strong sunshine and low night tem- 
peratures causes expansion and contraction of walls and roofs. 
In such situations it is best to avoid as far as possible the mixture 
of building materials with different coefficients of expansion. 
In Europe the range of surface temperature may be as much 
as ioo° F., and in dry sub-tropical countries such as Iraq and 
Iran and in the south-western United States even more. The 
coefficient of linear expansion a is given by the relation l t = 
l (/-fa/), where l is the length of a bar of the material at a 
temperature of 0° C. and l t is the length at a higher tempera- 
ture t° C. Approximate values of the coefficient a for tem- 
peratures between o° and 20° C. (32° and 68° F.) expressed 
in units of io~ 6 (-oooooi) are: — 



Iron and steel . 


. 


10 to 12 


Cement and concrete 


10 to 14 


Brick 


. 


9 to 10 


Sandstone 


7 to 12 


Glass 


. 


About 9 







A coefficient of 1 o in these units means that a bar of the material 
increases by 1/10,000 of its length with a rise of temperature by 
10° C. (18° F.). An iron girder 100 feet long at 32° F. would 
be about half an inch longer at 100° F. Wooden beams do not 
increase appreciably in length on heating, but their cross- 
section increases slightly. The coefficients of expansion of wood 
(in the unit used above) are 3 to 5 along the grain and 34 to 60 



SITING AND DESIGN OF HOUSES AND FACTORIES 67 

across the grain. But the effect of heat on wood is masked by 
the much greater effect of humidity. 

Duration of bright sunshine. — In view of the importance of 
sunshine not only for heating buildings, but for many other 
aspects of life, Table 10 has been constructed to give the 
average duration of bright sunshine in each month in hours 
per day. "Bright" sunshine is defined as sunshine sufficiently 
powerful to operate a Campbell-Stokes sunshine recorder, in 
which the sun's rays are focused by a glass sphere on to a strip 
of card, where they leave a charred trace as the sun moves 
across the sky. The sun can burn the card even when shining 
through thin cloud, but it must be bright enough to cast a clear 
shadow. On the other hand, the sun does not burn the card 
when it is near the horizon; the limiting range is generally 
3-5 °, but in the smoky atmosphere of large towns sunshine 
may not be recorded unless the sun is io° or more above the 
horizon. 

Most of the averages in Table 10 were obtained with 
Campbell-Stokes recorders, some by other instruments. 
They have been extracted from the official publications of 
the various Meteorological Services. Where instrumental 
records of sunshine are not available the duration can 
be estimated from observations of cloud amount by a method 
designed by C. E. P. Brooks (1929). 

Loss of heat in cold weather. — In cold climates the main problem 
is to prevent excessive loss of heat in winter. The extent of this 
loss is well brought out by the comparison of observations of 
temperature on the roof of Victory House, Kingsway, London, 
and in the nearby open space of Kensington Palace (W. A. L. 
Marshall, 1948). On the average of a complete year the roof 
was only about 1 ° F. warmer than the open space, but the 
difference was much greater in winter and at night. The night 
minima in January and February averaged 3 F. warmer on 
the roof than in the open, and in periods of severe frost the 
difference might be as great as 8-1 o° F., though part of this 
difference was due to the natural increase of temperature 
upwards on cold clear nights. Similar results were found at 
Debrecen, Hungary, by D. Berenyi (1948). The greater part 
of the loss of heat is due to conduction to the outside air. 

This depends on : ( 1 ) the tivickness of the walls and roof; 
(2) the insulating property of the materials; (3) the surface 



68 



CLIMATE IN EVERYDAY LIFE 







to ^ to o r-» 


tO CD 




^ 




^ 




oco 


o 


o 




to 




CO 


>< 




CO tJ< CO tJ- <N 


T^T^ 




-* 




m 




Tf Tf 


^ 


^ 




^ 




■* 


o 




w to T^ N w 


r-» t}< 




CO 




CO 




O Tf 


Tf 


m 




to 




<* 


u 




IH HH M -I O 


*"" M 








M 




H* l-l 


o 


M 




o 




° 


> 




t^ co -coco 


CO M 




o 




r^ 




CD r^ 


<N 


CO 




CO 




CO 


£ 




W ^ <N M O 


a CM 




CM 




CM 




M M 


M 


e< 




o 




o 


o 

o 




co m cm o 9 


m m 




to 




CO 




to iO 


i^ 


M 




CO 




co 




N COCOMN 


-^ CO 




CO 




TJ* 




CM CO 


<N 


CO 




C4 




CM 


Oh 




►h CD COCO ■«*« 


to CD 




rh 




t^. 




mto 


CO 


Tf 




o 




M 


13 

CO 




tJ< ^ r^ rj- CO 


iO m 




m 




to 




^ m 


^ 


T}H 




m 




^ 


bb 




a C^t^C^I^ 


lOCO 




m 




a 




O CO 


to 


O 




m 




to 


3 
< 




iO iT) -^ iO CO 


to t*» 




to 




CD 




r-. r^ 


m 


m 




to 




to 


^ 




lO l> CO CO tJ< 


Tf Tf 




to 




to 




CD •* 


w 


m 




M 




to 


3 




into mto ■* 


to CO 




to 




CD 




to CO 


i^ 


m 




CD 




CD 


2 




CD m co r^ e» 


cm o 




r~^ 




to 




cd m 


CD 


CN 




M 




CO 


3 
►-3 




m r~-to to in 


IO CO 




to 




CO 




in r* 


r^ 


to 




CD 




CO 


>* 




r^» r^ co mto 


c,tp 




CD 




a 




Tfir^ 


m 


o 




C4 




w 


% 




lOtO mtO ^ 


CO r- 




to 




i^ 




to r^ 


r^ 


to 




CO 




CO 


< 




T^x^r^ o co 


HH (-» 




CO 




i> 




cm to 


(N 


"*■ 




(N 




o 




rt* lO "^ lO CO 


iO iO 




m 




m 




in in 


m 


m 




to 




m 


o 




CD to iO CO -<j* 


CO HI 




CD 




M 




* CO 


o 


CO 




o 




m 


a 

2 




CM CO CO CO IN 


^Tf 




co 




^ 




CO ■<*• 


CO 


CO 




^3- 




CO 


JO 




CD m iO m o 


t^co 




CD 




CD 




CO o 


to 


to 




CD 




CO 






►h W <M 0) M 


CO CM 




CM 




cm 




« CO 


1-1 


CM 




M 




M 


d 




tJ< r^ m Th -f 


CM CD 




r-» 




CD 




co r^ 


CD 


1^ 




w 




CO 


CT3 




_,_,_,►_, O 


CM - 












HH l-l 


O 






1-1 




o 












• 




• 




• ' 


• 


• 




• 




• 








, . 




, 




. 




. . 


. 


. 




. 




, 






^-^ 




























u 


1 

5 


1 •**& 

^u- ss a y 


U 

i 2g 
-s § 

•2SC 
^5 


Li 
g i 

1§>C0 


1 


1 

Q 


V 

s il 

«-l <H 

PQC--. 


a 

V 

bp 

i rt 

il 

ii 


a 

1*1 
t5 


1 g 


4 


1 

1 



SITING AND DESIGN OF HOUSES AND FACTORIES 69 



(JlNO)H 


rf r^co O 


CO 


CO 


•* 


CD rj* m 


CO 


CO 


~ co 


m 


Tfrr^ tJ< ■* 


Th> CO ■* m 


i^ 


Tt« 


m 


mco t^ 


'f 


Tf 


m ■*■ 


r~- 


Tf 10 m m 


mco 10 w 


_ 


co 


co 


CD m CN 


r^ 


CO 


CN CD 


CO 


M ijHM M 


M - ~ 


-*■ 


1-1 


" 


CN CO M 


O 





« 


■* 


m CjO <n<o 


cd co cn cd 


CO 


_, 


Th 


m T^co 


CN 


CO 


- CD 


CO 


« tKNh 


M M CJ M 


T^ 


CN 


CN 


CO^H 






<N M 


Th 


CD COCO O 


co r-« 10 >-> 


CD 


CO 


in 


"*< <N CO 


CO 





„ t^co 


Tf 


COCO CO CO 


CO <N CO ^ 


CO 


CO 


T}< 


m n co 


CN 


CO 


CO CO 


CO 


^-0(0 ^ 


O CO - O 


CN 


M 


CO 


tOO « 


_ 


m 


CD m 


CO 


ioco m^ 


10 rf m<o 


CD 


10 


CO 


co t-* ■* 


n 


m 


m n 


CO 


O ""f tJ<(£> 


COCO cdco 


M 





CO 


M O ~ 


r- 


,_, 


n r— 


M 


co O r^to 


co mco t^ 


~ 


CO 


CO 


CD O CD 


CO 


r^ 


I^CO 


1-1 


« t^r-o 


mco r>» 


CD 


CO 


m 


O <N >-< 


m 


CO 


r~- <n 


_, 


coi^^ 


t^. tOf^CO 




CO 


CD 


CD m CD 


CO 


CO 


co r^ 


w 


CNCO ^CO 


CO CN <N CO 


CO 


CO 


_, 


•*■ m 


r^ 


co 


m cd 


_ 


r-» CDr^CO 


r~-co r^ r- 


O 


CO 


CD 


t^ CD t^ 


CO 


CD 


CD CO 





Tf tO O CO 


CO m coco 


m 





m 


(N CO « 


CD 


CO 


- CO 





toco t^co 


r^co r- r~- 


CO 


r^ 


CO 


t^ r-co 


CO 


CO 


CO CO 


CD 


^H(£l M 


r^i rf CN CO 


CO 


t^- 





H C5N 


M 


CD 


CD O 


O 


mr- m m 


mmiom 


r^ 


10 


CO 


co in co 


CO 


m 


m m 


CO 



cococjt}* in o o co cm cp co in o cd co cd oh Tf 

•^co rhco co co t^ ^h co co Tt< in in cn co co t^tj- co 



r«* co co yi^ cd o cocn ^ 

com^ftN ct h ei co m cn t^-^co >- cm n « m 



cp r~- o 00 t^ m cdco co co cd cor-" >-> co 7^ ^° °? 

M TfClH •—•*—••—••—• T^ w M CO CO CN M M -CM Tf 



ho 



bii pill U 31 £1 i§i*i i.& My il 



70 



CLIMATE IN EVERYDAY LIFE 





u 




CO 


01 O 




en 




cj co m 




co en m 




o 


M 


^co •- 




m 




^ 




co 


t^CO 




^ 




mco «*• 




« M CO 




CO 


CO 


m m m 




-* 




CJ 

Q 




_ 


m co 




x^ 




m en cj 




o - r*. 




CO 


M 


COCO J^ 




CO 






CJ 


Tf Tf 




O 




H COM 




O O CO 




** 


m 


m co -^ 




CJ 




> 

o 

55 




CO 


ci r- 




CJ 




CJ coco 




MM« 




en 


r^ 


co co en 




CO 






CO 


iO rf 




M 




CJ CO « 




O OCO 




CO 


m 


co m Tf 




CO 









in 


CO CO 




en 




Th r- (n 




coco t> 




en 


CO 


o mtN 




in 






m 


CO CO 




CJ 




Tf -f CO 




O O CO 




-* 


r- 


r^co m 




«* 




qL 




CI 


COCO 




^ 




'tO CO 




m co t}* 




CJ 


M 


co m co 




<* 








CO 


r-co 




m 




CO MO 




m cj r- 




r- 


en 


co co m 




CO 




be 
< 




m 


<*■ Tf 




en 




m en t^ 




m r- in 




m 


CJ 


CO CD CJ 




M 






o 


en m 




co 




co co r- 




CJ coco 




en 


~ 


CD I^N 




r-» 




>. 




i^ 


m tJ* 




o 




co "tr- 




O CO i-i 




in 


o 


co en en 




CO 




3 




o 


O <N 




en 




ee en r-~ 




^ <* m 




o 


CJ 


CO i^co 




r^ 








co 


m -> 




-* 




en rf cj 




oo « 




■* 


CO 


m cj m 




CO 




3 




en 


en -• 




en 




x^co r^ 




CO tJHO 




en 


o 


mco >* 




m 








cj 


I s - rf 




r^ 




hco m 




•-I co en 




^ 


CO 


w mr-> 




M 




S 




CO 


co en 




CO 




r-«CO CO 




co coco 




CO 


en 


mco m 




<* 




"C 




en 


r«.co 




Tj< 




m <n « 




cj >«f m 




CO 


co 


en enco 




CO 




ft 
< 




CO 


t^co 




CO 




mco in 




CJ cjco 




co 


CO 


CO T^ Tj- 




CO 




u 




en 


CJ o 




M 




co o h 




■* en « 




r^ 


CO 


•"" "*" CO 




CO 




ra 

s 




Tj« 


CO !*■ 




TjH 




tJ-CO Th 




H M b 




<* 


CO 


co m rj« 




CI 




J2 




I s -. 


CJ -^ 




CO 




r^ co o 




CO TfCO 




-* 


r-» 


t4h coco 




m 




ft 




co 


coco 




CJ 




co m co 




h 3CO 




co 


m 


CO ■'t CO 




CJ 








r^ 


CO M 




CO 




m o m 




>*HI M 




CO 


m 


r- Th en 




(S. 




•—5 




CJ 


■* m 




M 




CJ <* M 




O O CO 




M 


in 


•* Tf- CO 




CJ 










• • 




• 








• • • 




• 


• 






' 










. . 




• 




. . . 




. . . 




. 


• 


. . . 




• 




U 

u 

a 

ft 


4-> 
1 CO 

s 

s y 

1 3 

OS 


Spain — 
Barcelona 
Madrid 


1 

1 

CO 


| 

"o 

o 

O 

t/5 


4 

e 

-3 

• > 

1 


Berne 

Lugano 

Zurich 


U.S.S.R.— 
Leningrad 
Moscow . 
Vladivostok 


Li 

|# 

O V 


1 

1 


China — 
Hongkong 
Nanking . 
Shanghai . 


1 

| 

1 


a 

M 

O 

H 



SITING AND DESIGN OF HOUSES AND FACTORIES 7 1 



m •-« cd m eo 




rh- 


COCO 









M 




co 




CD 




iO 


CD CO 


co 




CD 


co r^r-co co 




<* 


10 <o 




t^ 




CO 




r^ 




CD 




r-» 


CD O 


CO 




r^ 


•^ CD t*» CDOO 




CO 


CO CD 




r^ 




CO 




CO 




r^ 




<N 


COCO 


t*» 




CO 


co r^ cd t^ 10 




CO 


CO iO 




10 




-*■ 




iO 




CO 




CO 


co r*» 


CO 




CO 


00 mo n 







(NCO 




ct 




<N 




CD 




CD 




O 


rfCO 


CI 




CO 


CDCO cnco -*■ 




lO 


iO iO 




CO 




iO 




iO 




r^ 




iO 


COCO 


r- 




r^ 


cd co w ■* 







<N CO 




iO 




"f 




O 




CD 




r». 


O O 


M 




CO 


co to co r-« 10 




CO 


CO Tj* 




t^ 




lO 




1^ 




CD 




r^ 


O O 


t^ 




CD 


CDCO tOCO <N 




M 


M CO 




<N 




^ 




iO 




<*• 




CO 


M M 


10 




t^ 


r^ m 10 Tf 10 




CO 


CO TjH 




r^ 




m 




r-» 




« 




CD 


M M 


^ 




CD 


t^CO (M CDCO 




M 


>-. t>» 




M 




Th 




to 




iO 




_ 


C* <N 


CO 




CO 


^ Th 1 CO CO ^ 




tO 


co to 




r-» 




r^ 




i^ 




ej 




CI 


<N Ci 


CO 




CD 


CO •-< CO -^co 




CD 


CO CO 




CD 




a 




^ 




M 




iO 


lOCO 


CO 




CO 


lO «f <N CO CO 




iO 


r^.co 




iO 




r^ 




r^ 




CO 




M 


<N Ci 


CO 




CO 


coco co m CO 




CO 


lO M 




CO 




CO 




CO 




M 




CO 


rhCD 







tO 


t^. m to Th to 




iO 


Tf IO 




t^ 




iO 




co 




CO 







ci a 


tO 




CO 


O MOOIM 




t^- 


CO tJ< 




O 




M 




<N 




CO 




CO 


tJ^CO 


to 




_ 


OOD ONN 




tO 


COCO 




CO 




CO 




CO 




» 




CO 


1-1 t-l 


<* 




CO 


CO 0* lOO O 




CO 


f a 




_ 




01 




_, 




CO 







t^co 


tO 




CO 


01000 C) 




a 


CO co 




CO 




CO 




CO 









r-» 





CO 




co 


CD to coco 




10 


r-O 




tO 




<N 




CO 




CD 




Th 


0* r^ 


CO 




w 


CD O O l^CO 




M 


lO CO 




r~» 




CO 




t^ 




CO 




LO 


CD CD 


t^. 




r^ 


HiOM CDCO 







N O 




M 




M 




CO 




M 




CO 


l^ t}< 


(N 




CO 


O CD r^co 




<N 


^co 




t^ 




1^ 




r^ 




r^ 




CO 


r^co 


CO 




CO 


CO tOCO t-» CD 




CO 


r~- M 




_ 




O 




CO 




tH 




CO 


CD to 


CO 




to 


co co CDr^co 




<N 


coco 




CO 




CO 




CO 




CO 




«f 


CO I** 


CD 




CO 








• • 




• 




• 




• 




• 




• 


• • 


• 




• 






• 


• • 




• 




• 




• 




• 




• 


• • 


• 




• 


Allahabad 
Bangalore 
Bombay . 
Calcutta . 
Madras 


1 
§ 

1 


G 

.a 
1 


Japan- 
Nagasaki 
Tokio 


1 


a. 

g 


u 
u 

I 8, 

§, 5P 


1 

• S 
a 

1 


d 


1 

1 
3 


q 
B 

1 




Egypt- 
Alexandria 
Cairo 


IS 

1 ° 

Is 


1 

q 

1 
5 


u 

3 
.O 



72 



CLIMATE IN EVERYDAY LIFE 



Si 

03 


b 






p> in r*» coco 
io io »o to uo 




OWOCOWCOCDT^OOp 

cb f^cb cb f^ f» in r^cb cb r» 


CD 

in 


cp in 

cb r^ 


o 

V 

Q 


6 




b 


rf T}- hH LO in 

« CI « h N 




tJ< r» cd in ^fcp o w o rfin 
mcoinr^'^fT^ci ■* i -<t , inTh 


p 


in w 

cb a> 


i 

55 


en 
b 




CI 

cd 


COCO CO «co 
co cm ci a cn 




coco t-» r^ tjh m mco r- cn co 
r-» rh cb r^cb in ci in incb in 


cp 


T 9* 

cb cb 


•4-> 

o 




b 




p> 


O T^op p a 
in tj« Th ^f ^ 




r~-cb f< f- r^-cb ^cb t** r^cb 


m 


cp cp 
cb r^- 




cd 




CO 

cb 


CO O CO CD O) 

cb cb cb cb in 




cp cdco Lncp p cp co Ln p r^. 
cb rococo r^-cb cb cb cDCDt>> 


CO 


pep 
cb t^ 


bb 

3 


CD 
CO 




CD 


CO O) CI CO ^f 

cb t-cb cb cb 




iOCf)OC<(N'*Hp^rt , 9 

cb cd cd cd f^cb cd cd b cb cb 


CI 


01 CD 

incb 


•-3 


cb 




CD 


cp Tf cn p> CO 

CDCO CD CD CD 




co cp co cp co co« i-< Thcp p 

CO O CDCDr^CDO O — CDCD 


cp 


CO CO 

co in 


| 
•-3 


o 
b 




CO 


CO mCO O CO 

cb cb cb cb cb 




O p)^M pCO ^ w pip CO 

bbbbbia>cDb««cD 


CO 


in t^ 

co in 




b 




cb 


OlO WN h 

r^- r^ f~- f^cb 




r-> « r--cp w co in w t^ r-^cp 

CD CDCO CD CDCO F* CD CD CDCO 


^ 
■* 


Thcb 




CO 

b 






CO CO CO coco 

f^cb cb cb cb 




tj< p> co io in ci p p>cp ci p> 
cd r^cb o cb cb r^ r-»cb cd t*~ 


cp 
cb 


cp t*. 

cb cb 


•s 

i 


b 




T* 1 

cb 


*hOCO* 

io in in ^ in 




cp ci «n in p p co ^<cp in in 
f-^cb r^ cd t^ f- r^cb cb r^cb 


CO 

cb 


in ^" 


JO 

V 


CO 

b 




CO 

cd 


cd coco o r«* 

CO ^ CO CO ^ 




cn cp r^. ci r^cp cp cp r-» •* w 
t^ in I-* cdco cb co in incb cb 


to 

cb 


in rh- 

cb cb 


d 

o3 


b 






<N CO <N M CO 




cp n cocn cor^N ^^« in 
in r^cb r~» in^j* ci -^ -^ in ^ 


p> 

cb 


ci pi 

CDCO 


1) 
o 


1 1 


1 c 

<; 03 

4 


d 

llftgi'S 


3 




1 * 

k 

a 03 

- PP 




1 • • -d 

M> b-» X 

-• . . 5 O .0.9 c3 
jH^3 Q <u C u 5 ♦S'3 03^ 


Ih • 

1 <! 03 

S O O 

•Sea 

55 j3 O 



SITING AND DESIGN OF HOUSES AND FACTORIES 73 



10 o co o 9 ri< yhcp m r-cp incoN 

r-» <b <b t^ r- t^-sD ioto ^to io m m 



to 


CO 

b 


CO 


9 


to 


r^ w rj< 9 cn 9 ci 
cdco t>^^6^ 


CD rh 1 to 

cb to £•» 


b 


CD 


M CO 

io r>. 


01 

to 


9 
cd 


CD 


r-^ 


CD 

CO 


cp cn to ^ r-tp co 
cb cb £"»*«» f*» CD ir^ 


co r^f^ 


9 


io 


to to 




CO 


to 

to 


CO 


CO 

CO 


CO CO tO CN t^CO CO 

Neb r^tb cb t^ r^ 


CO « CO 

ibeb cb 


CO 


to 


iO tN 

to to 


co 
cb 


to 
in 


CD 

to 


CO 
CO 


CN 


cn 9 co cp t*» CD •* 

cb cb to m toco t^ 


to to iO 


to 




o " 
cb ^ 



CO 

cb 


CN 

tO 


cp 






CN r^-to m CD o « 
io £"» tO tO ^ftb I s * 


CDCO tO 

^ ^h ^ 


CO 
tO 


CO 


CD CO 

cb cb 


CD 


tO 


CD 


9 


CO 


t}- cp to cn cn co 9 
•^ftb Th ^ ^ totb 


CD 9 to 

CO ^f CO 


tO 

to 


CO 


O CD 

b n 


cp 
b 


^ 


CO 

to 


tO 


to 


^tb 


CN CD r^CO CO 
tJh CO CO ^ tO 


CD CN tO 
CO CO CO 


CO 

cb 


CO 
io 


cp to 

CO N 


o 
cn 


•O 


cp 
to 


CD 
tO 


to 


cp to 
■*tb 


m co toto r*» 

tO "^ -^ tO iO 


CN CO tO 
^ -^ "* 


n 


N 


« 9 
cb r- 


9 




tO 


CO 

to 


7^ 
cb 


9 9 ^N CO CO CN 

cb f^tb *f to f^tb 


CD CO CN 

^ ^ io 


tO 

io 


cb 




tO 


CO 
CD 


o 


co 

CO 


tO 


no 9 Thtp r>. to 
n r-- f»tb to cb to 


CD ^f Tf 

to totb 


to 

CN 


co 


a to 
cb cb 


co 


9 


7f 


to 


cp 


CO tO CD O TftNCN 

cd f-cb nco b> r» 


N N CD 

to to to 


CD 


to 


coco 
io r- 



tO t^ 

4^ b 



o toto to CO -f CO 


coco CO 


to 


r^ 


CD CD 


o r^ n Noo o ro- 


r^to r^ 


o 


to 


Tt-r- 




iisjeg^s- •Ills' ,| gjs ||i 

Mlsiiii tsis if Si 511 

^ ^ ^ £ s 



74 CLIMATE IN EVERYDAY LIFE 

area per unit volume of interior space ; (4) the temperature and 
wind-speed of the outside air. 

(1) Other things being equal, the insulating property of 
a wall is proportional to its thickness. Hence, with ordinary 
building materials the thickness of the walls must be greater 
the lower the winter temperature. While in southern England, 
Belgium and Holland a thickness of 9 inches generally gives 
sufficient insulation, 10 inches is the minimum in Western 
Germany, 15 inches in Central and Eastern Germany, 20 inches 
in Lithuania and Poland, and 28-30 inches in Russia. A rough 
rule is that the thickness should be 9 inches if the mean tem- 
perature of the coldest month is 34 F. or more, and should 
increase by 1 inch for every degree by which the mean tem- 
perature of the coldest month falls below 34 F. In windy 
situations, however, the thickness needs to be increased, hence 
the generally thick walls of upland farmhouses. 

(2) For equal thickness the lower the heat conductivity of 
the material the better the insulation. Air is one of the best 
insulators provided that free circulation is prevented; hence 
dry, porous materials are better than close-grained stone ; metal 
is the worst of all. The following figures give approximately the 
heat required to maintain a temperature difference of 50 F. 
between the inside and outside of a wall 10 inches thick, ex- 
pressed in B.T.U. per hour per square foot of wall area: — 





Material of Wall 




Air Space (Still air) 

Felt 

Dry Sandstone 


o*8 Asbestos 
1 -3 Cement 
1-9 Iron . 


4 

10 
1600 



The insulating power of porous material is destroyed if the 
pores are filled with water, so that a waterproofed surface is 
necessary, especially on the weather side. Loss of heat through 
saturated walls is many times (5-20) that through dry, porous 
walls. Wet walls are also cooled by evaporation, and for this 
reason also it is desirable that walls should shed rain-water 
quickly instead of absorbing it. Double walls separated by an 
air space require provision against free circulation of air, other- 
wise the space will be filled with cold air which will cool and 
damp the internal walls, and ties must be arranged so as not 
to lead moisture inwards. 

A great deal of heat is lost through windows. This loss can 



SITING AND DESIGN OF HOUSES AND FACTORIES 75 

be minimised by making windows small as well as tight-fitting; 
if large windows are required to give good lighting in climates 
with very cold winters they should be double, separated by an 
air-space. 

In most buildings the hottest air collects under the roofs and 
ceilings, and as the wind speed is also greatest at roof level the 
roof is the greatest source of heat loss. Hence insulation of the 
roofs, which is often neglected, is especially important. 

(3) The loss of heat depends on the surface area, which 
depends on the square of the linear dimensions, while the 
volume depends on the cube. Hence, to a first approximation 
doubling the linear scale of a building halves the heat required 
for heating per cubic yard of interior space. However, as the 
wind speed increases with height above the ground (see p. 34), 
this gain is partly counterbalanced by increased loss from the 
upper floors. The increase of wind speed with height is greatest 
at night, when the air is coldest, so that the upper rooms of 
high buildings in winter tend to be cold in the morning. The 
most economical arrangement for heating is probably long, low 
buildings with well-insulated roofs, arranged parallel with the 
prevailing wind, but with some form of wind-break to prevent 
the uninterrupted sweep of the winds between the buildings. 

(4)(jStill air conducts heat very slowly, but the loss of heat is 
greatly increased by even a moderate wind. Outdoor air is very 
rarely completely calm; even if there is no natural breeze, con- 
vection from walls and roofs of buildings introduces some move- 
ment. As it passes a large building the air is warmed slightly, 
but the more rapid the movement the smaller the warming. 
The cooling effect is roughly proportional to the square root 
of the wind speed. In cold air with a wind of 12 m.p.h. the 
loss of heat is about three times the loss at the same temperature 
in calm air. Allowance has to be made for the different tem- 
peratures of winds from different directions ; the coldest winds 
in winter are, generally speaking, those which have had the 
longest passage over land, and come from higher latitudes. In 
London, for example, in December to February, the mean 
temperature of east and north-east winds which come from 
high latitudes in Europe is about 30 F. Winds from north, 
which blow mostly over cold oceans, have a mean temperature 
of about 34 , and those from north-west about 38 . Westerly 
winds which came originally from high latitudes but have had 



76 CLIMATE IN EVERYDAY LIFE 

a long passage over the relatively warm Atlantic have a mean 
of about 42 °, and south-westerly winds reach 48 and provide 
warm, muggy conditions. On the other hand, the cold, dry 
south-east winds which blow across France are much colder, 
averaging only 39 F. The temperature depends on the origin 
of the air as well as the direction of the wind (J. E. Belasco, 
1945)- 

LIGHTING 

The unit of illumination is the lux, which is defined as the 
direct illumination on a point one metre from a point source 
of light of one candle-power. One lux equals 0*093 foot-candles. 
In the open, except in conditions of thick fog or very heavy 
cloud, there is always enough illumination so long as the sun 
is above the horizon. On cloudless days the direct illumination 
from the sun well above the horizon exceeds 10,000 lux. In 
moderately cloudy weather the illumination is almost directly 
proportional to the height of the sun, and amounts to about 

Table ii. — Average length of day, sunrise to sunset, hours 

and tenths. 





Jan. 


Feb. 


March 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Latitude 


























66° N. 


4'4 


8-o 


11 


5 


15-2 


19-0 


22-9 


20-6 


16-8 


13-1 


9-6 


5-8 


29 


64° 


5-4 


8-4 


11 


6 


14-9 


i8-i 


20-5 


19-5 


i6- 3 


13-0 


9-8 


6-5 


4-3 


62° 


6-i 


8-7 


11 


6 


14-7 


17-5 


19-4 


18-6 


15-9 


12-9 


io-o 


7-o 


5-2 


6o° 


67 


90 


11 


7 


14-5 


17-0 


i8-6 


17-9 


15-6 


12-9 


io-i 


7-5 


5-9 


5 8° 


7-2 


9-3 


11 


7 


i4'3 


i6-6 


17-9 


17-3 


15-2 


12-8 


10-3 


7-9 


6-5 


56° 


7-6 


9-5 


11 


8 


14-1 


16-2 


17-4 


16-9 


15-0 


127 


10-4 


8-2 


7-o 


54° 


8-o 


9-7 


11 


8 


13-9 


15-9 


17 -o 


16-5 


14-8 


127 


10-5 


8-5 


7-4 


52° 


8-3 


9-8 


11 


8 


13-8 


15-6 


i6-6 


16a 


14-6 


127 


io-6 


8-8 


7-8 


5o° 


8-6 


io-o 


11 


8 


13-7 


15-3 


16-2 


15-8 


14-4 


12-6 


io-7 


9-0 


8-i 


45° 


9-1 


10-3 


11 


8 


13-4 


14-8 


15-5 


15-2 


14*0 


12-5 


10-9 


9-5 


8-8 


40 


9-6 


io-6 


11 


8 


132 


14-3 


14-9 


147 


13-7 


12-4 


ii-i 


io-o 


9-3 


35° 


io-o 


10-9 


11 


9 


13-0 


139 


14-4 


14-2 


13-4 


12-4 


n-3 


10-3 


9-8 


30° 


10-4 


ii'i 


11 


9 


12-8 


13-6 


14-0 


13-8 


13-2 


12-3 


n-4 


io-6 


IO-2 


25° 


107 


n-3 


12 





12-7 


13-3 


i3'6 


13-5 


13-0 


12-3 


"•5 


10-9 


io-6 


20° 


ii-i 


"•5 


12 





12-6 


13-1 


13-3 


13-2 


12-8 


12-3 


117 


II-2 


10-9 


15° 


n-3 


n-6 


12 





12-5 


12-8 


13-0 


12-9 


12-6 


12-2 


n-8 


H-4 


11-2 


IO° 


n-6 


n-8 


12 


1 


12-3 


12-6 


12-7 


12-6 


12-5 


12-2 


n-9 


117 


"■5 


5°N. 


n-9 


12-0 


12 


1 


I2'2 


12-3 


12-4 


12-4 


12-3 


12-2 


120 


n-9 


n-8 


o° 


I2-I 


I2-I 


12 


1 


I2-I 


I2'I 


I2-I 


I2-I 


I2-I 


I2-I 


I2-I 


I2-I 


12-1 


5°S. 


12-4 


12-3 


12 


1 


I20 


n-9 


n-8 


119 


n-9 


I2-I 


12-2 


12-3 


12-4 


IO° 


12-6 


12-4 


12 


2 


n-9 


n-7 


"•5 


n-6 


n-8 


12-0 


12-3 


12-6 


12-7 


15° 


12-9 


12-6 


12 


2 


n-7 


n-4 


II '8 


n-3 


n-6 


12-0 


I2'4 


128 


13-0 


20° 


13-2 


12-7 


12 


2 


n-6 


11*2 


IO-9 


II-O 


n-4 


12.0 


12-5 


130 


133 


25° 


13-5 


12-9 


12 


2 


"•5 


10*9 


106 


107 


II-2 


n-9 


12-6 


13-3 


136 


30° 


13-8 


13-1 


12 


2 


H'S 


io-6 


IO*2 


10-4 


II-O 


n-9 


12-7 


136 


14-0 


35° 


14-2 


I3'3 


12 


3 


11*2 


103 


9-8 


IO'O 


108 


II-8 


12-9 


139 


144 


40 


14-6 


13-6 


12 


3 


II-O 


9-9 


9-3 


9-6 


105 


II-8 


13-1 


I4'3 


149 


45° 


151 


13-9 


12 


4 


io-8 


9-5 


8-8 


9-1 


IO-2 


117 


13-3 


147 


155 


50° 


15-8 


I4'3 


12 


4 


io-6 


9-0 


8-i 


8-5 


9-8 


ix-6 


i3'5 


15-2 


16-2 


52° 


i6-i 


14-4 


12 


5 


10-5 


87 


7-8 


8-2 


97 


n-6 


136 


15-5 


16-5 


54° 


16-4 


14-6 


12 


5 


103 


8-4 


7-4 


7-8 


9-5 


"•5 


13-7 


15-7 


16-9 


56° 


i6-8 


14-9 


12 


5 


IO-2 


8-i 


7-o 


7-5 


9-3 


n-5 


13 9 


161 


i7'4 


5 8° 


17-2 


151 


12 


6 


io-i 


7-8 


6-5 


7-0 


9-0 


n-5 


140 


i6- 4 


17-9 


6o°S. 


17-8 


15-3 


126 


9-9 


7-4 


5-9 


6-5 


8-8 


114 


14-2 


16-9 


186 



SITING AND DESIGN OF HOUSES AND FACTORIES 77 

700 lux for every io° of solar altitude. The effect of cloud or 
obstacles is felt chiefly before sunrise and after sunset. 

Civil twilight covers the periods before sunrise and after sunset 
during which there is enough light for ordinary outdoor occupa- 
tions. In Britain it officially begins and ends when the sun is 
6° below the horizon, and its duration in different latitudes 
according to this definition is tabulated in the abridged edition 
of the Nautical Almanac (London, H.M.S.O., annually). Its 
duration depends on the angle which the path of the sun makes 
with the horizon; it is shorter in low than in high latitudes, 
and in all latitudes it is shortest at the equinoxes and longest 
at the solstices. 

The duration of evening twilight in different latitudes of the 
northern hemisphere is as follows : — 



Latitude 


Duration of Evening Twilight, minutes. 




March 21 


June 21 


Sept. 21 


Dec. 21 




30 
40 

50 

54 
58 
62 
66 


21 

,24 
27 

32 

35 
39 
44 
51 


22 
27 
32 
44 

54 
1 h. 16 m. 


21 

24 
28 

33 
36 
40 
46 

53 


22 
27 
30 
38 

43 

5 1 
1 h. 5 m. 

1 h. 40 m. 



North of 6o° N. twilight at midsummer lasts all night. In 
practice the duration of twilight depends also on cloud con- 
ditions and on the obstruction of the horizon, particularly 
towards the rising or setting sun. In an open situation with 
a clear sky it is possible to read out of doors until the sun is 
about 6|° below the horizon. 

Fig. 10 shows, on the left, the illumination in an open space 
with different distances of the sun below the horizon on days of 
clear sky, overcast or foggy days, and in heavy rain ; and, on 
the right the effect of buildings on either side, as in a street. In 
a street running east and west the twilight illumination is about 
15 per cent, greater, and in a street running north and south 
about 20 per cent, less tiian the average shown in the figure. 



78 



CLIMATE IN EVERYDAY LIFE 



A snow cover on the ground increases the illumination by about 
20 per cent. 

Inside a building the illumination is very much less than in 
the open, the ratio depending on the size of the windows, the 
direction in which they face, the "skyline" visible from the 



400 



300 



bi200 



IOO 



SO 



1 


1 A-Open 
1 B-Sky Iine30 - 
I C-Sky line 45* 
I D-Sky line 60* 
\ E-Skyline75* 


\ 


lA Clear sky 

n 


\ 


WW 


v\ 


eVvsX 


^^\^^^^ 


x ^^^^__ 



Fig. 



2 345 601 2 345 

Depression of sun below horizon, degrees 

0. — Effect of cloud and sky-line on illumination. 



400 



300 



200 



100 



so 



windows, and, of course, the position in the room. This point 
is discussed in Chapter XI. 



PURITY OF AIR 



<A good supply of clean air is essential both for living and 
working conditions. Natural ventilation depends on the wind, 
but wind also carries impurities from sources of pollution, such 
as chimneys, and raises dust. Dust is most troublesome near 
ground level, especially at corners, while solid particles and 
gases due to combustion are most abundant at the general roof 
level. The air is in general cleanest at about three-quarters of 
the height from the ground to the general roof level, but on 
the lee side of buildings eddies are apt to carry pollution down- 
wards. If clean air is necessary for manufacturing processes, 



SITING AND DESIGN OF HOUSES AND FACTORIES 79 

as with such products as confectionery and cosmetics, the site 
should be to windward of the main sources of pollution.: In 
Britain this means on the south-west side of large towns; 
the Bourneville works, for example, are to the south-west of 
Birmingham. Even in such situations, however, the air is con- 
taminated when the wind blows from the town. Vegetation is 
very effective in cleaning air, and a belt of trees a short distance 
to windward (or in the direction of sources of pollution) is a 
valuable precaution. Atmospheric pollution is discussed further 
in Chapter IX. 

Atmospheric impurities have a damaging effect on the 
structure of buildings. A great deal of gaseous acid impurity 
is brought down by rain, and if the rain penetrates the walls 
it carries the impurities with it and the acids attack the lime 
in the walls. When the walls dry out these salts are brought to 
the surface and form a hard crust. Where this crust is broken, 
holes develop in the softened material below. To avoid this 
trouble it is necessary to get the water away from the walls 
quickly by means of a smooth, impervious covering. Atmo- 
spheric impurities also attack metals; iron, for example, rusts 
more quickly in the presence of sulphur dioxide than in pure 
air, and still more quickly when solid carbonaceous particles 
are present (see Chapter IX). 



LOCAL SITUATION IN REGARD TO FROST, RAIN, SNOW, ETC. 

Frost and fog. — The local distribution of frost and fog in 
relation to topography was discussed in Chapter I. A com- 
bination of frost and fog is bad from all points of view — health, 
expense of lighting and heating, damage to buildings, clothes 
and products in course of manufacture, and delay in transport, 
The best situation is in the middle third of a gentle southerly 
slope from a narrow ridge, with a good air drainage, but below 
the strong winds and cloud of the hill tops. A site on the side 
of a plateau is less favourable because there is a greater supply 
of cold air from the more extensive surface. Frost and fog 
develop at night in hollows and in valleys in which the natural 
air drainage is obstructed. The obstruction may be a narrowing 
of the valley, a railway embankment or even a belt of trees, 
and there is likely to be a sharp frost and fog line at or a little 
above the top of the obstruction. Hence, if a site in such a 



80 CLIMATE IN EVERYDAY LIFE 

valley is under consideration a survey of such obstacles to free 
air flow should be made. 

A striking example of the effect of a narrow valley in creating 
a frost-hollow, aided to some extent by a railway embankment, 
is described by E. L. Hawke (1944) from near Rickmansworth 
in Hertfordshire. He sums up by quoting figures to show that 
the night climate of this low-level Hertfordshire valley is almost 
identical with that of Braemar, at a height of 1,110 feet in 
Aberdeenshire, and Moor House, 1,840 feet up in the northern 
Pennines. A still more remarkable and instructive example 
from Austria is quoted by Hawke (1946) in a chatty article 
on frost-hollows in general. 

Even on slopes situations differ as regards liability to frost. 
The cold air flowing downwards from the hills follows the line 
of least resistance and is readily diverted by obstacles. These 
may be so placed as to concentrate the air flow in well-defined 
channels, where on clear nights ground and buildings are 
exposed to a strong, cold breeze. Such situations can usually 
be recognised from a study of the ground, and it may be possible 
to divert the air by planting trees or building walls obliquely 
across the path of the air. 

The disadvantages of hollows are mainly limited to the night 
and early morning; it is only during severe cold spells in winter 
that frost and fog persist all day. In sunny weather hollows 
are bright and warm during the day, though the sun soon goes 
off. Hence, such a hollow may not be unsuitable for buildings 
such as offices which are not in use at night, for the cold of 
early morning is partly counterbalanced by shelter against 
wind and driving rain. Frost hollows are bad sites for resi- 
dences, gardens and orchards. 

Frost damages walls and roads by "exfoliation" or flaking 
off of the surface. The important factor is not so much the 
degree of cold as the frequency with which the temperature 
oscillates about the freezing-point. This can be calculated from 
the difference between the number of days of screen frost [i.e. 
days on which the air temperature in the shade falls below 
32 ° F. at night) and the number of ice-days, or days on which 
the temperature does not rise above freezing-point at any time. 
These figures can be obtained from the Meteorological Service. 
In Britain ice-days are rare; data collected by the Meteoro- 
logical Office show that the average number is only one or two 



SITING AND DESIGN OF HOUSES AND FACTORIES 8 1 

a winter near the coast, rising to four at moderate heights inland 
in England and seven in southern Scotland; on higher ground 
the number is greater. Frost days, on the other hand, vary from 
about ten on the coasts of Cornwall and Wales to 40 or 50 over 
most of England (70 in the lowlands of the eastern interior) 
and 100 over much of Scotland. The number of frost days 
depends very much on the local situation. In colder regions, 
such as central and eastern Europe, Siberia and much of North 
America, ice-days are very frequent in winter, and the risk of 
frost damage probably does not differ much from that in 
Brijain. 

^An important factor in design is the depth to which frost 
penetrates in winter. This depends on three factors: (1) the 
duration and severity of the period with mean temperature 
below 32 F.; (2) the presence and thickness of a snow cover; 
and (3) the nature of the soil. (1) The depth of penetration of 
frost may be taken as roughly proportional to the square root of 
the accumulated temperature below freezing-point. The latter 
figure can be found by taking all months with mean tempera- 
tures below 32 ° F. and adding up the deficits. For practical 
purposes, since the duration of the winter is roughly propor- 
tional to its severity, the depth of frost penetration in bare 
ground may be taken as very roughly one to two feet when the 
temperature of the coldest month is about 5 F. below freezing- 
point. In the great frost of 1895, for example, the mean air 
temperature of the period from 26th January to 19th February 
was about 26 F. in London, and the lowest readings of the 
earth thermometers at 1 foot depth ranged from 28 to 32 °. 
Water mains were frozen at considerably greater depths, 
generally 2-3 feet and even more, but this was attributed to 
the fact that, close by, the pipes were near the surface so that 
the water was already practically at freezing-point, and that 
heat conduction along iron pipes lowered the temperature of 
the mains even further. 

/ In countries with cold winters frost shortens the period during 
which building can be carried on owing to its effect on cement 
and mortaft> In central Europe the risk of frost is practically 
limited to the period with mean daily temperature below 
50 F., but urgent work can be carried on by the use of special 
mixtures or by electrical heating. According to B. Hrudicka 
(1937-8) the energy needed for the latter is one Kw.hour per 



82 CLIMATE IN EVERYDAY LIFE 

cubic metre for each Centigrade degree by which the air 
temperature falls below o° G. or 32 ° F. 

Days on which the temperature falls below 32 ° F. are termed 
"ice-days" or "frost-days." In an ice-day frost persists all day. 
In a frost-day which is not an ice-day temperature rises above 
freezing-point at some time during the day. In Britain it is 
estimated that following a morning with frost, on the average 
about one-third of the working day between 7 a.m. and 5 p.m. 
has a temperature below 32 F. Consequently the loss of 
working days due to frost is about one-third of the number of 
frost-days, less allowance for Sundays and holidays. 

(2) Loose snow is a good insulator, and a thick cover of snow 
prevents the temperature of the ground beneath from falling 
much below 32 ° F. The conductivity of new snow is three to 
four times that of air, roughly half that of ordinary compacted 
soil, and equal to or slightly greater than that of loose soil or 
loose sand. Thus the insulating effect of a layer of loose, new 
snow 1 foot deep is roughly equivalent to that of 2 feet of dry 
sub-soil and to a much greater thickness of rock. As the snow 
becomes compacted, however, its thermal conductivity in- 
creases, and for old snow it is three or four times that of new 
snow. 

(3) The insulating effect of soil depends almost entirely on 
the amount of air which it holds per unit volume. The con- 
ductivity rises rapidly as this is replaced by water or ice. Hence 
the saturation of soil greatly increases the depth of freezing. 

When water freezes it expands, and a serious effect of the 
freezing of wet ground is the "heaving" of the soil, which may 
cause cracks in buildings. To avoid this it is necessary to extend 
the foundations below the depth to which the soil is likely to 
freeze. 

Rainfall increases with height above sea-level, especially on 
windward slopes. In Britain the increase is at the rate of 2 or 
3 per cent, per hundred feet in the moderately hilly country of 
the south and midlands, and 4 per cent, in mountain regions 
such as north Wales. For example, the average rainfall is 
24-5 inches a year at Camden Square (110 feet) and 26-2 
inches at Golder's Hill Park (350 feet), an increase of nearly 
3 per cent, per hundred feet. Except in very hilly country, 
however, these local differences of rainfall are not of great 
importance. In some parts of the world there is a very great 



SITING AND DESIGN OF HOUSES AND FACTORIES 83 

difference between the windward and leeward sides of the hills, 
and if the wind blows steadily from the same direction this fact 
is of great economic importance. In tropical regions like 
Hawaii or Southern India the rainfall may exceed 100 inches 
a year on the windward slopes of the hills fronting the ocean, 
and fall to only about 30 inches a short distance away on the 
leeward side. The former is too wet for settlement, the latter 
too dry, unless water can be obtained from across the summit. 

Rain causes loss of time in building operations. The duration 
of rainfall in Britain is measured as the time during which rain 
is falling at the rate of 0-004 inch (o-i mm.) or more per hour, 
and averages about 600 hours a year in the lowlands of the 
west and north, and 500 hours a year in the Midlands and 
London. Corresponding figures are difficult to obtain for other 
countries. In the temperate regions between about latitudes 
45 and 6o° N. the duration of rainfall may be taken as roughly 
proportional to the number of raindays. In lower latitudes the 
rainfall is heavier and the duration for die same amount of rain 
is correspondingly less. 

As a guide to the effect of rainfall on ordinary outdoor occu- 
pations the following scale of rainfall intensities is suggested 
(amounts in inches) : — 

Table 13. — Scale of rainfall intensity. 



Intensity 


5 mins. 


30 mins. 


1 hour 


2 hours 


1 day 


1. Barely perceptible 


— 


below 0-002 


below 0-004 


below 004 


below 0-005 


2. Very light 


below 0-004 


0-002-0-01 


0-004-0-015 


0-004-0-02 


0-005-0-04 


3. Light 


0-004-0 -oi 


0-01-0-03 


0-015-0-04 


0-02-0-05 


-04-0 • 1 


4. Moderate . 


0-01-0-03 


0-03-0-1 


0-04-0-15 


0-05-0-2 


0-1-0-4 


5. Moderately heavy 


0-03-0-08 


0-1-0-3 


0-15-0-4 


0-2-0-5 


0-4-1-0 


6. Heavy 


0-08-0-2 


0-3-0-75 


0-4-1-0 


0-5-1-25 


1-0-3-0 


7. Very heavy 


0-2-0-4 


Q-75-1'5 


I O-2-O 


125-25 


3-0-6-0 


8. Torrential 


0-4-0-8 


1-5-2-5 


2-0-3-5 


2-5-5-° 


6-0-12-0 


9. Phenomenal 


above o-8 


above 2-5 


above 35 


above 5-0 


above 12-0 



Falls up to and including "Light" do not interfere with 
ordinary outdoor activities; in Britain these account for about 
a quarter of the total duration of rain. "Moderate" falls in- 
terfere with but do not necessarily prevent outdoor work. 
"Moderately heavy" falls prevent outside activities, "Very 
heavy" falls cause minor flooding and "Torrential" falls cause 



84 CLIMATE IN EVERYDAY LIFE 

extensive flooding. But the effect of a given fall varies very 
much according to the normal rainfall of the district. Thus 
10 inches in a day has never been recorded in Britain (9*56 
inches was recorded at Bruton, Somerset, on 28th June 191 7), 
but at Cherrapunji, India, a fall of 10 inches would pass without 
notice. Extreme rainfalls are discussed in Chapter X. 

The terminal velocity of raindrops falling in still air is approxi- 
mately as follows : — 



Diameter of drop, inch . 


. 0-02 


0-05 


O'lO 


0-15 


0-20 


03 


Velocity, feet/second 


• 13 


16 


24 


28 


30 


30 



Raindrops cannot grow to a greater diameter than 0-3 inch 
because, owing to the resistance of the air, at that size they 
flatten out and break up into smaller drops. The inclined 
velocity of a drop carried by the wind is the square root of the 
sum of squares of wind speed and terminal velocity. 

Snow is important in building design in certain regions be- 
cause of the additional load which it places on roofs. The snowi- 
est parts of the world are the mountain region from Alaska 
to the northern United States, especially British Columbia 
and north-eastern Japan; any mountain region facing winds 
off the sea in a cold climate is liable tojhe avy falls of s now. The 
weight of freshly fallen snow varies according to its "fluffmess"; 
an average figure is about 6J lbs. per cubic foot. As the snow 
becomes compacted its weight per cubic foot increases, and in 
old snow may be as much as 30 lbs. In the coastal ranges of 
western North America the weight of snow on level surfaces 
may be as much as 250 lbs. per square foot, roughly a ton per 
square yard. In these regions houses are always built with 
steep roofs, which allow the snow to slide off; a slope of 6o° 
does not carry snow. In the north-eastern United States and 
eastern Canada the snow cover on flat roofs may reach a weight 
of 50 lbs. per square foot; in New York it is customary to allow 
for 40 lbs. In Britain and western Europe the maximum depth 
of undrifted snow on level ground rarely exceeds 2 feet, equiva- 
lent to about 13 lbs. per square foot. As, however, owing to 
drifting the depth of snow may be locally greater it would seem 
desirable to allow for a snow load of at least 20 lbs. per square 
foot on large flat roofs. Where the local topography causes the 
snow to pile up in drifts the thickness may be much greater, 
and hill country is especially liable to deep drifts. Two or three 



SITING AND DESIGN OF HOUSES AND FACTORIES 85 

feet of snow may seem a trivial addition to the weight of a roof, 
but it has sufficed to cause the collapse of large buildings. A 
notable example was the Knickerbocker theatre disaster in 
Washington in 1922. 

WIND 

"TWind speed and direction must be taken into consideration 
both in the siting and construction of all buildings. Especially 
important are wind pressure and the effect of the wind in dis- 
tributing smoke and noxious fumes. To appreciate the effects 




Fig. 1 1 . — Illustration of turbulence. 

of wind it is necessary to understand something of wind struc- 
ture. With very few exceptions wind does not consist of a 
regular flow of air like a smoothly flowing river. Over open, 
fairly level country it is made up of a series of whirls or eddies, 
which travel along with the general stream of air. Whenever an 
eddy passes, the wind at any point changes in direction or 
speed, usually both. 

In Fig. 11 the straight arrows represent the general wind, 
with speed V. The circle represents an eddy blowing in an 
anti-clockwise direction with speed W. As the eddy passes over 
point A the two speeds V and W are added together, and the 
wind momentarily blows with speed V-fW. This is a gust. At 
the point B, on the other hand, the two speeds will partly cancel 
out, and the wind speed will momentarily drop to V— W; this 
is a lull. At G the wind speed does not change much, but its 



86 CLIMATE IN EVERYDAY LIFE 

direction swings first to the left and then to the right. Since the 
distribution of eddies is quite irregular and they completely 
fill the wind stream, the result at any one place is a continuous 
succession of gusts and lulls and small changes of direction. 
The diameter of an eddy is generally between 50 and 1 00 feet, 
increasing with the speed of the general current of air, so that 
eddies, and consequently gusts and lulls, follow one another at 
intervals of a few seconds, averaging about seven seconds. A 
wind stream made up of such eddies is said to be turbulent. 

The wind recorder in use at official stations in Britain, the 
Dines Pressure-tube anemometer, is able to follow these rapid 
fluctuations so that the trace of the recording pen shows a broad 
band of nearly vertical lines. The instrument possesses some 
inertia, of course, and the gust velocities which it records are 
averages over about two seconds. An example is shown in 
Fig. 12, reproduced by courtesy of the Air Ministry. 

The tops of the velocity lines represent gusts and the bottoms 
lulls. The width of the band of lines, or the range of wind 
speed between gusts and lulls is a measure of the gustiness of 
the wind. Other examples representing a variety of exposures 
are given by E. Gold (1936). 

The gustiness with winds from the same direction at any one 
place is roughly proportional to the average wind speed and is 
measured by the "gustiness factor" G, which is the ratio of 
the average difference between the gusts and lulls to the average 
mean speed. For example, in a very open situation at Spurn 
Head a typical record shows an average wind velocity of 
25 m.p.h. with gusts of 30 m.p.h. and lulls of 20 m.p.h. The 
gustiness factor is 10/25 or °'4- O n tne other hand, an anemo- 
meter exposed on the roof of the Science Museum, London, 
with an average wind speed of 20 m.p.h. had gusts of 40 m.p.h. 
and lulls falling almost to calm, a gustiness factor of 2. At a 
height of 33 feet the gustiness factor is 0-3 or 0-4 on flat islands 
and about 0-5 on the coast with winds blowing off the sea; 
inland it is greater, depending on the nature of the country. 
Probably an average value of 0*5-1-0 would be appropriate for 
good open situations, rising to 1 -5 in sites surrounded by trees 
or high buildings. The factor decreases with height, i.e. the 
wind becomes smoother the higher it is above the ground. It 
is greatest in the afternoon and least in the early morning, when 
it is only half to two-thirds as great. 



SITING AND DESIGN OF HOUSES AND FACTORIES 



87 



Time On 31 d 09k 17m Off 1 d ooh 10 m Marksat h m GMT Sheet Hot 

MID- 
NIGHT 

230 12 3456 7 8 9 10 MB. 

—100,14 



Holyhead 



January \* 194-3 



il 1 



it 



li 






90' 



80 



70 



60. 



-50- 



8 .3 



4< 
o 

=> 
</> 

_ w 

2 2 

»- 
r 
< 



23 1 2 3 

~ T_ ' r N T ~ 



^Jl^ . | 



350- 
E300 = 



H 1£- 



JH L 



200- 
150g 

feiooS 

: o 

E 50 



Fig. 12. — Specimen of anemogr am. 



88 CLIMATE IN EVERYDAY LIFE 

The range of wind direction over short intervals of time also 
depends on the gustiness. With a small gustiness factor the 
wind varies only a few degrees ; the extreme range over an hour 
might be between SW. and WSW. With a high gustiness factor 
the range of direction is much greater, between say S. and 
WNW. In a very turbulent situation, such as the lee of a row of 
houses, the wind may blow from all round the compass. 

An anemometer records only the horizontal wind, but there 
are similar variations in the vertical movement. This can 
easily be seen by watching the smoke from a factory chimney, 
which starts as a thin stream, expanding downwind both side- 
ways and up-and-down, to form a cone with horizontal axis, 
until the smoke becomes too thin to be seen. 

In addition to the constant succession of gusts and lulls, on 
some days there are occasional periods of a few minutes when 
the wind rises to a peak velocity in about ten seconds and falls 
off again, generally more gradually over a minute or more; 
the peak velocity lasts for only about two seconds. These are 
convectional eddies and may occur simultaneously along a line 
hundreds of feet wide. They are known as squalls and may do 
a great deal of damage. These are discussed in Chapter X. 

The general structure of the wind described above is modified 
when it meets obstacles such a^steep hills or buildings. 

Wind pressure on buildings. ^r-When the wind encounters a 
building it presses on the windward face and exerts a force 
directed towards the interior of the building. On the leeward 
face the air is sucked away by the wind currents passing over- 
head and on each side, and a partial vacuum is formed which 
exercises a suction effect; on this side therefore the force on 
the walls is directed outwards. 

The air piled up against the windward face escapes by passing 
over the roof; with isolated buildings the air also flows past on 
either side. Owing to its momentum the air blowing up the 
face of the building cannot flatten out immediately, and there 
is also a suction or lifting effect on a part or the whole of the 
roof. If the roof is flat or has a pitch of less than 45 ° the lifting 
effect extends over the whole roof. If the pitch exceeds 45 ° 
there is pressure on the windward and suction on the leeward 
side. Suction is especially powerful at the eaves of a nearly 
flat roof. In fact, over a building as a whole suction exceeds 
pressure ; windows are more often blown out than inV> 



SITING AND DESIGN OF HOUSES AND FACTORIES 89 

The most disturbed region is just to leeward of the ridge of 
the roof, where an eddy forms with a horizontal axis, the air 
moving in the direction of the wind above roof-level and 
against the wind lower down. But owing to the succession of 
gusts and lulls the velocity is always changing, and a succession 
of eddies breaks off and drifts away down-wind. These eddies 
cause alternating pressure and strong suction on the leeward 
side of the building, which is therefore most liable to damage. 

The pressure exerted by the wind is proportional to the 
square of the speed; on a flat plate normal to the wind it is 
usually taken as P=o-oo3V 2 where P is pressure in lbs. per 
square foot x and V is velocity in miles per hour. A. M . Thomas 
(1930) terms this the "velocity pressure." In the rear of the 
plate there is a suction effect which must be added to the 
pressure on the face. Consequently the total force acting on 
the plate exceeds the velocity pressure. On a rectangular flat 
plate of great length, such as a bridge girder, the total force is 
twice the velocity pressure. On a tall building directly facing the 
wind Thomas gives it as i-6 times the velocity pressure, while 
on a square, flat plate such as a signboard the ratio falls to i • i . 

Curved surfaces allow the air to flow past with less hindrance, 
consequently the pressure on them is less. For cylinders the 
effective area is not half the total surface, but only the area 
projected on a plane at right angles to the wind, or slightly less 
than one-third of the total surface, and the total force is only 
about o-8 (chimneys, standpipes) or 0-7 (water tanks) times 
the velocity pressure on this effective area. 

The failure of a gasometer in the tropics was found by C. A. 
Middleton (1937) to be due to suction causing the plates on the 
lee side to bulge outwards. The rivets were corroded much 
more on the lee side than to windward, and under the vibrations 
caused by continual variation of the suction with the passage 
of gusts and lulls they tore out. There was also suction on die 
dome of the gasometer. Middleton found both by calculation 
and experiment that the total suction on a cylinder in a high 
wind is much greater than the total pressure. Pressure is 
exerted on only one-sixth of the circumference directly facing 
the wind. The greatest suction is on either side, slightly to 
windward of the diameter normal to the wind. 

1 In Fig. 12 wind pressure is shown in millibars (inb.). One mb. is equivalent 
to slightly over 2 lb./sq. ft. 



90 CLIMATE IN EVERYDAY LIFE 

On surfaces inclined to the wind the pressure is naturally less 
than on similar surfaces directly facing the wind. Duchemin's 
formula, quoted by R. Fleming (1930), gives the factor by 
which the wind pressure on a surface at right angles to the wind 
must be multiplied as 2 sin 0/(i-f-sin 2 6), where 6 is the angle 
between the surface of the building and the wind. 

Building regulations in different countries differ as to the 
wind pressure to be allowed for in construction, but generally 
they require a safety limit of 30 lbs. per square foot, either over 
the whole structure or over the exposed parts. This is equivalent 
to a maximum gust velocity of 100 m.p.h., which is only ex- 
ceeded, and that rarely, in the windiest parts of the British 
Isles (see below). It should be remarked that the maximum 
gusts are limited both in area and duration. Since the diameter 
of an eddy is about 50 to 100 feet, it follows that the maximum 
average pressure over any structure 100 feet or more long is 
appreciably less than the maximum at any one point. An 
estimate of 75 per cent, is quoted by Fleming as a safe figure. 
Also light structures such as radio towers, stack pipes, etc., 
respond quickly to wind pressure, and for these the maximum 
gust velocity over two seconds gives a good measure of the 
stresses. Heavier buildings, however, have a greater inertia, 
and for these the effective maximum wind speed to be con- 
sidered is that over a period of minutes. This is only about 70 
per cent, of the maximum gust velocity; the exact figure 
depends on the gustiness factor, being greatest when the latter 
is least. 

Shelter by other buildings generally reduces the forces on the 
building sheltered, but may bring the latter entirely within the 
area of suction to leeward of the exposed building. Such cases 
are probably rare. Solid walls and fences have a similar effect. 
The best shelter is provided by a belt of trees or an open fence, 
which reduce the wind speed without creating dangerous 
eddies (see Chapter XII). Dangerous lifting forces on a roof 
may be minimised by the use of a longitudinal opening under 
the ridge of the roof, such as a louver, which often forms part 
ofjhe system of ventilation. 

/One possibility of damage which is rarely considered is that 
the period of oscillation of an erection may coincide with the 
periodicity of gusts, so that resonance occurs. K. Doring (1925) 
gives the average period of oscillation of a chimney of reinforced 



SITING AND DESIGN OF HOUSES AND FACTORIES 9 1 

concrete 330 feet high as 2-4 seconds, but it is possible that with 
stronger winds the period increases. R. Fleming (1930) gives 
recordings indicating that buildings of light construction have 
an oscillation period of four seconds in gales, while those of 
heavier structure have a smaller oscillation of shorter period. 
The average interval between gusts being about seven seconds, 
it appears that danger from this cause will not often arise. 

Maximum gust velocities. — The most violent known winds at the 
earth's surface occur in tornadoes. These are very impressive, 
estimates up to 400 m.p.h. having been quoted, but the strongest 
winds occur over a very small area, and the risk to any par- 
ticular building is small. Even in the worst areas severe damage 
at any particular spot is only likely to occur once in 1,000 to 
2,000 years. Tornadoes have therefore been left out of account 
here, since it is not economically practicable to construct 
buildings which will stand up to the centre of a tornado. They 
are discussed further in Chapter X. 

Dissipation of smoke.^—A study of the wind direction and 
structure is especially important where chimneys emitting 
smoke and noxious fumes are concerned. There are two 
aspects : the influence of topography on the local wind direc- 
tion and the effect of turbulence in spreading and dissipating 
smoke and gases> 

The local wmcls tend to follow the valleys, especially if these 
are narrow and steep-sided. Strong, gusty winds may sweep 
up and down such valleys. This fact must be considered in 
siting works which emit noxious fumes to avoid risk of damage 
and claims for compensation in other parts of the valley. A 
notable example occurred in the Columbia River valley, where 
fumes from smelting works at Trail in British Columbia travelled 
southwards into the State of Washington, causing damage to 
crops. The method recommended to mitigate the trouble in 
this case is described in Chapter IX. 

Besides disturbances of the horizontal direction the topo- 
graphy may give the wind a prevailing upward or downward 
direction. The general air flow must follow the main features 
of the slope of the ground, but after blowing up a hillside the 
upward direction is continued by momentum for a short 
distance. After that the air descends again, and generally 
reaches the ground again at a distance of roughly a quarter to 
half a mile to leeward of the hill crest. In such a situation, with 



92 CLIMATE IN EVERYDAY LIFE 

a strong wind blowing over the crest, there is an almost per- 
manent down draft. When the leeward slope of the hills is very 
steep (or behind a block of tall buildings) there is a powerful 
eddy, the wind actually blowing against the prevailing direction. 
Such down drafts may be very troublesome, interfering with 
combustion of furnaces or blowing the smoke down to the 
ground. Between this down draft and the crest there is a 
relatively calm region in which, however, the wind is very gusty. 

Even in the absence of such down drafts, however, the gases 
from a chimney eventually reach the ground. It is only neces- 
sary to watch the plume of smoke from a chimney to see that 
it expands down wind, forming a cone with a horizontal axis. 
If the wind is very gusty the plume expands rapidly, and the 
cone has a broad apex. If the air flow is unusually smooth the 
plume expands slowly and the cone has a narrow apex. If the 
draught in the chimney is strong the gases are thrown some 
distance above the top before they begin to spread, but this 
effect is not appreciable unless the upward velocity of the 
issuing gases is twice the wind speed (slightly less than twice 
with strong winds). For the purposes of calculation it is more 
or less cancelled out by the down draft immediately behind the 
top of the chimney, so that the smoke trail may be assumed to 
start at chimney height. The rate of spreading is also affected 
by the diameter of the chimney, which governs the size of the 
trailing eddies with vertical axes formed behind the chimney; 
the turbulence is greater behind a broad chimney than behind 
a narrow one. The smoke from an isolated chimney will there- 
fore reach the ground at a distance from the foot of the chimney 
which depends on the height and breadth of the chimney, the 
gustiness of the wind, and to a small extent the draught in the 
chimney. According to O. G. Sutton (1947) the maximum 
concentration at ground level from a chimney 160 feet high 
occurs at a distance of 720 yards in a very gusty wind, 1,300 
yards under average conditions, and 3,400 yards when the air 
is nearly free from turbulence. These figures apply to normal 
horizontal flow of the air; if there is a down draft the maximum 
concentration of smoke is greater and occurs nearer the foot 
of the chimney. 

If the chimney rises from the roof of a factory or other 
building, or if there are neighbouring buildings to windward, 
conditions are more complicated. Behind the building there is 



SITING AND DESIGN OF HOUSES AND FACTORIES 93 

a region of very turbulent air, which may extend to well above 
the level of the roof of the building. "Down wash" begins as 
soon as the smoke reaches this turbulent layer. R. H. Sherlock 
and E. A. Stalker (1940) carried out a series of experiments on 
models at Chicago from which they concluded that the down 
wash of stack gases occurs in two stages: (1) Down flow in 
eddies in the wake of the stack; (2) Dispersion of gases within 
the turbulent mass of air behind and above the buildings. In 
their experiment the building supporting the chimneys pre- 
sented a blunt face to the prevailing wind, so that the turbulent 
layer extended nearly to the top of the chimneys. The height 
of the latter was 250 feet, and the model indicated that gases 
reached the ground at a distance of only 1,250 feet. 

The nearer the top of the turbulent layer extends to the top 
of the chimney the greater is the danger of serious contamina- 
tion of the air. If the effective height of the chimney can be 
raised above the top of the turbulent layer by a height greater 
than the diameter of the chimney there is little risk of serious 
contamination. Some of the gases will reach the ground at a 
greater distance by the ordinary operation of diffusion, but not 
in sufficient concentration to be harmful. 

The remedy adopted by Sherlock and Stalker was to fit 
nozzles to the chimneys of greater height than the diameter of 
the latter. This had three results: it increased the effective 
height of the chimneys ; it increased the upward velocity of the 
gases; and the eddies behind the nozzles being smaller than 
those behind the chimneys, the smoke was carried downwards 
less rapidly. These precautions were effective. This paper by 
Sherlock and Stalker is of great interest. One deduction is that 
in designing a factory, if smoke and noxious gases are likely to 
be troublesome, the face presented to the prevailing wind could 
with advantage be "streamlined" to some extent by setting 
back the upper stories and sloping the roof at a low angle. 

Ordinarily smoke spreads both upwards and downwards, 
and half of it is dissipated into the air above and rendered 
harmless. In certain weather situations, however, known as 
inversions, there is a cold layer of air near the ground with 
warmer air above. The top of the cold layer acts like a lid ; 
smoke accumulates below this "lid" and may form a dense 
pall — this is, in fact, the origin of town fogs. In such situations 
the emission of noxious gases may be very harmful, especially 



94 CLIMATE IN EVERYDAY LIFE 

in sheltered hollows and valleys. The possibilities in this direc- 
tion are illustrated by the fog in the Meuse valley on 3rd to 5th 
December 1930 in the industrial area between Huy and Seraing, 
when sixty-four people (all old or in poor health) died, as well 
as a number of cattle. 



RAIN AND WIND 

The penetration of rain into walls and through cracks, and 
direct mechanical damage by rain, are greatly increased when 
the rain is accompanied by a strong wind. The penetrating 
power of rain under these conditions is proportional to the 
intensity of the rain and the square of the wind velocity. The 
raindrops are broken up into a very fine mist, which is forced 
into the pores and cracks in the wall by the wind. This effect 
was discussed by W. Thelm (1933), who termed the occurrence 
" Schlagregen " (pelting rain, "shock rain"). The minimum 
conditions for damage by shock rain are a wind of 34 m.p.h. 
with a rainfall at the rate of o-oi inch per hour, or about a 
quarter of an inch a day. Moderately severe shock rain occurs 
with a wind speed of 25 m.p.h. and a rainfall of 0-5 inch an 
hour, a wind of 30 m.p.h. and a rainfall of 0-36 inch an hour, 
or a wind of 35 m.p.h. and a rainfall of 0-25 inch an hour. If 
these conditions are maintained more or less continuously for 
two or three days severe damage is likely to occur. Unfortu- 
nately it is difficult to obtain the necessary data to find out how 
frequent and widespread shock rains are, but they are probably 
liable to occur once or twice in most winters in exposed places 
on the western coast of Great Britain and the north-west coast 
of Europe. 



CHAPTER III 
"TEMPERATE" CLIMATES OF THE OLD WORLD 

THIS and the four following chapters give brief descrip- 
tions of the main climatic characteristics of the various 
land masses of the world. Their aim is not to set out 
strings of figures ; such statistics can be obtained from the tables 
in the Appendix or, if required in greater detail, from the 
Meteorological Office, but to present a simple picture of the 
main peculiarities and the progress of the seasons. The land 
areas may be divided into Polar, Temperate and Tropical, con- 
veniently separated by the Arctic and Antarctic Circles 
(66°3o' N. and S.) and the two Tropics (23°3o' N. and S.). In 
the polar regions the distinction between night and day almost 
vanishes except near the equinoxes ; the year consists of a very 
long, hard winter and a very short summer. In the "tem- 
perate" regions the changes from day to night and winter to 
summer are both important, but the range of temperature 
between the averages of the warmest and coldest months 
exceeds the average daily range of temperature. In the interiors 
of the continents the annual range is very great, and "tem- 
perate" is a misnomer; it is to be read as a synonym for "middle 
latitude." Between the tropics the temperature contrast of the 
seasons becomes less important, and is exceeded by the contrast 
between day and night. "Night is the winter of the tropics," 
but the real seasons are defined by the variations of rainfall. It 
so happens that the area in which the daily range of tem- 
perature exceeds the annual range is almost exactly limited by 
the tropics, the only exceptions being India, where it is bounded 
by a line from Bombay to Madras, and the western coastal 
regions of South America, where it extends as far south as 
Santiago. As the temperate regions are by far the most im- 
portant economically, these are discussed first. 

In accordance with usual meteorological practice the seasons 
are defined as (in the northern hemisphere) : winter, December 
to February; spring, Maich to May; summer, June to August; 
autumn, September to November. In the southern hemisphere 
these seasons are reversed, i.e. winter is June to August, etc. 

95 



96 CLIMATE IN EVERYDAY LIFE 

In high latitudes and in the interior of the continents these 
"seasons" are little more than conventions; the year consists 
of winter and summer, with short transitional periods of about 
a month. 

The weather of the temperate regions is governed by the 
constantly changing pressure distribution, in which the two 
main elements are the anticyclone and the barometric depres- 
sion. An anticyclone is a region of high pressure, light winds 
and generally dry weather. In summer the sky is usually clear 
and the weather sunny and warm, giving ideal holiday weather, 
but in winter anticyclones in western Europe the sky is often 
covered by a uniform layer of grey cloud. In winter also anti- 
cyclones frequently bring fog, especially in towns where the 
light winds cannot dissipate the smoke. In the interior of 
Siberia there is an almost permanent large anticyclone in 
winter; in other parts of the temperate regions anticyclones 
appear from time to time in all seasons and drift slowly along 
irregular tracks. 

The characteristic weather of most parts of the temperate 
regions results from the passage, in a general easterly direction, 
of a succession of barometric depressions. A depression is an 
area of low pressure round which (in the northern hemisphere) 
the winds blow in an anti-clockwise direction. In the southern 
hemisphere the direction of the winds is clockwise. In the 
northern hemisphere the winds in the foremost part of a depres- 
sion advancing from west to east are southerly or south- 
easterly and generally mild and humid, with moderate, steady 
rain. In the rear of the depression the winds are northerly or 
north-westerly and bring a change to colder weather with 
showers and fine intervals. As the depression passes away to 
the east the showers become less frequent and a day or two of 
fine weather precedes the arrival of the next depression. There 
are usually one or two breaks in the smooth circulation of the 
air on the side of a depression nearest to the Equator ; these are 
"fronts" where two masses of air of different origin come in 
contact. The "cold front," where the cold air in the rear of a 
depression replaces the warm air in the southern part, is usually 
marked by a rapid change of wind direction, with a squall and 
often a heavy shower. On the polar side of a depression the 
winds are easterly (north-east in the northern hemisphere, 
south-east in the southern). These winds are cold and often 



CLIMATES OF THE OLD WORLD 97 

bring heavy snow in winter. They are less liable to be inter- 
rupted by squalls than the winds on the equatorial side of the 
centre, and the snow or rain is often very persistent. 

Depressions vary greatly in size and intensity. The average 
diameter is 1,000 to 1,500 miles, but a deep winter depression 
may have a diameter up to 3,000 miles and bring heavy rain 
and persistent gales along the whole Atlantic coast of Europe. 
The wind may reach speeds of 60 or 70 m.p.h. averaged over 
an hour or more, and much higher speeds in gusts. Similar 
conditions are found in the Aleutians and on the western coast 
of Canada, Washington and Oregon, on the coast of New 
England, Newfoundland and the maritime provinces of Canada. 
In the interior of the continents the winds are less strong. In 
summer depressions are smaller and generally less intense, but 
severe storms sometimes occur and bring gales over limited areas. 

It is the frequent but irregular passage of depressions, and 
the intervening fine periods of varying lengths, the fresh but 
rarely destructive winds, the alternation of rain and sun, and 
the constant changes of temperature which give the climates of 
the temperate regions their stimulating character, especially in 
the coastal regions. The weather is often moderately uncom- 
fortable, but never so extreme as to numb the faculties and 
cause a feeling of hopelessness. The peoples of these regions are 
"conditioned" to change, and hence more alert and less liable 
to get into a rut than those of less variable climates. That is 
probably the main reason why the "cyclonic temperate" 
regions (see fig. 1) are among the main centres of progress. 

British isles (Appendix I — Birmingham, Cardiff, Dover, 
Falmouth, London, Manchester, Portsmouth, Tynemouth, 
Yarmouth ; Aberdeen, Edinburgh, Lerwick, Stornoway ; Belfast, 
Cork, Dublin). 

The climate of the British Isles is both equable and change- 
able. Extremes of heat and cold, or of drought and heavy rain 
are rare, but the weather seldom remains the same for longer 
than ten days or so at a time. The mean annual temperature 
below a height of about 500 feet is everywhere between 45 ° and 
52 ° F. On the west coasts and islands the coldest month is 
February, elsewhere January ; similarly, the warmest month in 
the west is August and in the east July. The difference between 
the warmest and coldest months nowhere exceeds 25 F., and 
on the west coasts and islands is only 15 . As is usual in oceanic 



98 CLIMATE IN EVERYDAY LIFE 

climates, winter tends to be prolonged into spring, which is 
a treacherous, uncertain season, and summer into autumn. 
Extreme temperatures are rare; in London in over a hundred 
years the thermometer has only once touched ioo° F. and has 
never fallen below 4 F. ; in an average year the range is from 
85°-io,°. Lower temperatures are found in the northern Mid- 
lands and southern Scotland, but in the more populous parts 
of the country it is not necessary to reckon with a temperature 
lower than 5 F. once in ten years. 

Really severe winters are rare; in the past seventy years the 
outstanding ones have been: 1879-80, 1890-91, 1894-5, 1939- 
40 and 1946-47. Of these 1890-91 and 1946-47 were the worst, 
of about equal severity. 

December 1879 was the coldest month of the century in 
France and central Europe, and the cold persisted into January; 
the Dutch waterways were frozen for nearly two months, and in 
Paris fifty people died of cold. In Britain the winter was not so 
severe, but deaths from cold were reported and evergreens were 
killed. On 4th December a temperature of — 23 F. was re- 
corded at Blackadder, in Berwickshire, but this is not officially 
recognised, as the thermometer was uncertified and not con- 
ventionally exposed. 

The winter of 1890-91 was remarkable for its long duration, 
from 25th November to 22nd January, rather than for the 
intensity of the frost. During this period the average tem- 
perature was below 32 ° F. over nearly the whole of England 
and Wales, and below 30 in East Anglia and the south-east 
Midlands. Skating in Regent's Park occurred on forty- three 
days, the thickness of the ice exceeding 9 inches, but the frost 
penetrated the ground to a depth of barely a foot. 

The frost of 1894-95 lasted from 30th December to 5th March 
with one break; the coldest day was nth February when 
— 1 7 F. was recorded at Braemar, the lowest in Britain which 
is officially recognised. From 9th to 17th February the whole 
of the Thames was more or less blocked by ice-floes, some of 
them 6-7 feet thick. 

The winter of 1939-40 was not so intense as that of 1894-95, 
but was longer and snowier. It was notable for the glazed frost 
at the end of January (see p. 47). 

The winter of 1946-47 ranks with that of 1890-91 as probably 
the worst in Britain since 1 789. Not only was it very long, but 



TEMPERATE CLIMATES OF THE OLD WORLD 



99 



there were heavy and repeated falls of snow, a long succession 
of sunless days and persistent biting east winds; the greater part 
of Great Britain and Northern Ireland was continuously snow- 
covered from 27th January to 13th March. Level depths ex- 
ceeded 2 feet and there was much drifting. The dislocation of 
road and rail traffic was unprecedented. 

The rainfall of Britain is generally moderate and well dis- 
tributed through the year. On low ground it increases from 
about 20-25 inches a year in south-east England to 40-50 
inches in the west and north-west. The wettest places are near 
the summits of the mountains; in Wales, Cumberland and 
western Scotland the rainfall approaches 200 inches. These 
rainy hilly regions are of value as gathering grounds for the 
water supply of large towns and for hydro-electric power. 

The rainfall is generally reliable, which is fortunate for the 
dense population of Britain with their large water requirements. 
Droughts lasting for more than six months are rare. The out- 
standing droughts since 1864 are summarised below. Fuller 
accounts of those from 1864-192 1 are given by C. E. P. Brooks 
and J. Glasspoole (1922). 





Table 14- 


-Droughts in Britain. 










Percentage of Average 


Rainfall 


Year 


Period 


TVTrtntVw 






1VXUI1LJ.10 


England and Wales 


Scotland 


1864 


April-August 


5 


61 


73 


1868 


May-July 


3 


38 


67 


1879-80 


October-January 


4 


36 


58 


1887 


February-July 


6 


57 


73 




February-October 


9 


68 


77 


1893 


March-June 


4 


43 


72 


1895 


February-June 


5 


62 


72 


1896 


January-May 


5 


60 


79 


1921 


February-July 


6 


49 


84 




February-October 


9 


58 


89 


1929 


January-April 


4 


47 


5i 


1938 


February-April 


3 


3i 


70 


1947 


August-December 


5 


57 


78 



Droughts are generally most severe in south-east England, 
where the rainfall is not only lower than in the rest of the 
country, but also more variable from year to year. 



100 CLIMATE IN EVERYDAY LIFE 

Evaporation is greatest in summer, when it generally exceeds 
the rainfall in the drier parts of the country, and in an average 
year most of the rain which falls in summer is evaporated with- 
out penetrating the sub-soil. The underground water, which 
maintains the springs and rivers in dry periods, is mostly the 
accumulation of the winter rains, so that a winter drought, 
though less spectacular than a hot, dry summer, has a greater 
effect on our reserves of wat^r. 

Although the winter weather of the British Isles can be 
described as stormy, widespread damage by wind is rare, and 
is chiefly limited to the blowing down of trees and to minor 
damage to buildings. There is some risk to trains travelling 
along exposed sections of line (the Tay Bridge disaster of 28th 
December 1879 was caused by a gale), and at Quilty in west 
Ireland an anemometer was installed which rings a bell when 
the wind reaches 65 m.p.h., as a warning to weight the trucks of 
trains, and again at 85 m.p.h., when traffic is stopped along an 
exposed section of the railway. The most severe storms generally 
bring heavy rain and may result in minor flooding. Serious 
floods may be due to a variety of causes : intense local thunder- 
storm rains, long-sustained heavy rains generally accompanying 
a slow-moving barometric depression, or a succession of storms 
with mild air and heavy rain following a period of deep snow 
cover and frozen ground. In tidal rivers the effects of floods 
are accentuated when the peak coincides with a period of un- 
usually high tide caused by the piling up of river water by 
storm winds. The flood in the Thames estuary on 6th~7th 
January 1928, when fourteen people were drowned by the 
flooding of basements in London, was due almost entirely to 
the piling up of the tidal water by a storm. For descriptions of 
the most severe floods, see C. E. P. Brooks and J. Glasspoole 
(1928). 

The storms which cause the greatest inconvenience in the 
British Isles are those in which strong winds are accompanied 
by heavy snow. These are popularly termed "blizzards," and 
though they do not bring the very low temperatures and dry, 
powdery snow of the true American blizzards, they are some- 
times not dissimilar. In the past seventy years we may recall 
three such storms in southern England : — 

iSth-20th January 1881. — This affected the greater part of 
England, and was most severe from Somerset to the Isle of 



101 

Wight; snow in the streets of London interfered with traffic 
for a fortnight. 

gth-i^th March 1891, which was worst in Devon and Corn- 
wall. Several trains were buried for days on Dartmoor, and 
passengers narrowly escaped starvation. 

26th December 1927. — This brought more than a foot of un- 
drifted snow on high ground, and the deep drifts piled up by 
the gale blocked some roads for a week. The snow was soft and 
clingy, and broke down many telephone wires. 

The worst snowstorms of recent years in northern England 
were those of the end of February and beginning of March 
1886 and 4th~5th March 1947. A detailed account, year by 
year, of snowfalls in the British Isles from 1876 to 1925 is given 
by L. C. W. Bonacina (1927). 

In spite of these occasional contretemps the climate of Britain 
and especially of south-east England is, on the whole, ideal for 
most forms of human activity. The average summer temperature 
is very near the optimum, and the winters are cold enough to 
be bracing, but not too cold for simple heating methods to be 
effective without elaborate air conditioning. The humidity is 
moderate, without either excessive dryness to cause nervous 
irritation and dust nuisance or long periods of enervating damp 
heat. The rainfall is adequate without being excessive. There 
is generally enough wind to clean the air and ventilate build- 
ings ; the dirty fogs of large towns are only isolated nuisances. 
Above all the weather is not monotonous. Northern England 
and Scotland are almost equally favourable, but the winters 
of the central and eastern districts are harsher and, especially 
in the west, the periods of favourable anticyclonic weather are 
fewer and shorter. The climate of Ireland is even more equable 
— in fact too equable, the winters being too mild to provide 
the tonic effect of this season in Britain. 

Although the succession of wet and dry, stormy and fine 
periods is at first sight chaotic, careful study has revealed a 
pattern. Around certain dates definite types of weather tend to 
recur in from half to almost all the years. An account of these 
"recurrences" was given by G. E. P. Brooks (1946c?). The 
principal tendencies are as follows : — 

(1) October to early February, stormy periods with minor 
anticyclonic interludes. 



102 



CLIMATE IN EVERYDAY LIFE 



(2) February to May, cold waves associated with north- 

easterly winds. 

(3) The summer period of alternating cool fresh north- 

westerly and warm, sultry south-westerly winds. 

(4) September and early October, spells of anticyclonic con- 

ditions and late "summers." 

The principal stages in the seasonal succession, with their 
average and peak dates of occurrence and the number of times 
they occurred in the fifty-two years 1 889-1 940 are as follows: — 



Type 



Early January, stormy . 
Mid-January, anticyclonic 
Late January, stormy 

Early February, anticyclonic 
Late February, cold spell 
Late February and early 
March, stormy . 

Mid-March, anticyclonic 
Late March, stormy 
Mid- April, stormy 

Late April, unsettled 
June, summer monsoon 
July, warm period 

Late August, stormy 

Early September, anticyclonic 

Mid- September, stormy 

Early October, stormy . 
Mid-October, anticyclonic 

Late October and early Nov 

ember, stormy . 
Mid-November, anticyclonic 
Late November and early Dec 

ember, stormy . 

Pre-Christmas, anticyclonic 
Post-Christmas, stormy . 



Average Dates of 



Beginning Ending 



Jan. 5 
Jan. 18 
Jan. 24 

Feb. 8 
Feb. 21 

Feb. 26 

Mar. 12 
Mar. 24 
April 10 

April 23 
June 1 
July 10 

Aug. 20 
Sept. 1 
Sept. 17 

Oct. 5 
Oct. 16 



Oct. 24 

Nov. 15 

Nov. 24 

Dec. 18 
Dec. 25 



Jan. 17 
Jan. 24 
Feb. 1 

Feb. 16 
Feb. 25 

Mar. 9 

Mar. 19 
Mar. 31 
April 15 

April 26 
June 21 
July 24 

Aug. 30 
Sept. 17 
Sept. 24 

Oct. 12 
Oct. 20 



Nov. 13 

Nov. 21 

Dec. 14 

Dec. 24 
Jan. 1 



Peak 



Jan. 8 

Jan. 20-21 

Jan. 31 

Feb. 13 
Feb. 22 

Mar. 1 

Mar. 13-14 
Mar. 28 

April 14 

April 25 



Aug. 28 
Sept. 10 
Sept. 20 

Oct. 8-9 
Oct. 19 

Oct. 2g\ 

Nov. 9, 12/ 
Nov. 18, 20 
Nov. 25 \ 
:-9J 



Dec. 



Dec. 19-21 
Dec. 28 



Frequency 
in 52 years 



45 

45 
44 

29 

22 

46 

27 
35 
37 

27 
(40) 



35 
43 
3i 

35 
35 

52 
34 
5i 

29 
43 



These characteristic spells of weather do not occur every 
year, and when they do come the dates are not always exact — 
there may be a range of a week on either side. They are there- 
fore in no sense forecasts of the weather to be expected in any 



CLIMATES OF THE OLD WORLD IO3 

particular year, but the table is a useful guide in planning 
activities when the date has to be decided more than five days 
or so ahead. Really "long range" forecasts, weeks or months 
ahead, are not yet practicable in temperate regions. In par- 
ticular, "weather cycles" such as the "sunspot cycle" of eleven 
years and the "Bruckner cycle" of thirty-five years are quite 
unreliable in spite of the publicity which they have received. 
This may be illustrated by the fact that the year 192 1, about 
the driest on record in England, came near the middle of the 
wet half of a Bruckner cycle. Official forecasts are usually 
reliable for twenty-four hours, and in general terms for forty- 
eight hours ahead, and are occasionally possible for five days 
or even a week, but anything beyond this is at present little 
more than guesswork. For a description of the organisation of 
weather forecasts in Britain, see E. G. Bilham (1947). For a 
detailed study of the climate of the British Isles, with numerous 
maps and tables, see E. G. Bilham (1938). 

north-west and gentral Europe (Appendix I — Austria, 
Innsbruck, Vienna; Belgium, Brussels; Czechoslovakia, Briinn, 
Karlsbad, Prague; Denmark, Copenhagen; Faeroes, Thorshavn; 
France, Bordeaux, Cherbourg, Paris, Strasbourg; Germany, 
Aachen, Berlin, Breslau, Frankfurt-am-Main, Hamburg, Leip- 
zig, Munich, Stuttgart; Holland, Flushing, Utrecht; Hungary, 
Budapest; Norway, Bergen, Oslo, Tromso; Sweden, Haparanda, 
Stockholm; Switzerland, Berne, Geneva, Zurich). 

Owing to the absence of north-south mountain ranges the 
climate of the whole North European plain between the Baltic 
and the Alps changes only gradually from west to east over the 
whole stretch from Britain to the Urals. The winters grow 
more severe as we pass eastwards (a very severe winter in 
London would be regarded as normal in Berlin), the snow 
cover is more regular and persistent, the rise of temperature in 
spring is more rapid, the summers are somewhat warmer, 
rainier and more thundery, and autumn to some extent loses 
its pleasant character as an extension of summer. The differ- 
ence is most marked in winter; summer in, for example, Warsaw 
is not greatly different from summer in London. The change 
is most rapid within a few miles of the coast; Paris, for example, 
in spite of its lower latitude, is on the average slightly colder in 
January than is London, and the extremes of cold are far more 
intense. This is because the moderating influence of the North 



104 CLIMATE IN EVERYDAY LIFE 

Sea on the cold north-east winds is absent. The lowest tem- 
peratures on record are (in ° F.) : Flushing, +3; Utrecht, —5; 
Berlin, —15; Warsaw, —22; Moscow, —31; Kasan, —34; 
Sverdlovsk, —43. The spells of mild, damp weather associated 
with stormy conditions in winter extend far into Europe, but 
become shorter and less pronounced as one goes eastwards; 
at the same time the frequency of gales decreases. 

In spring north-easterly winds reach their greatest fre- 
quency and cause spells of cold weather, which often 
damage crops. These spells are popularly associated with the 
famous "Ice Saints" of nth-i3th May, but actually such 
cold spells are liable to occur at any time during April and 
May. Apart from these cold spells spring is a pleasant, calm 
season. 

About the end of May there is often a sudden change in the 
prevailing type of weather. This brings in the "European 
Monsoon" (see p. 102) and takes the form of a rapid increase 
in the frequency and strength of westerly and north-westerly 
winds, especially on the Baltic coast of Germany. Fresh, breezy, 
showery weather sets in and usually lasts for about three weeks. 
By the end of June summer conditions of fine hot spells alter- 
nating with thundery rains have generally set in. Towards the 
end of September, in most years, especially in central and 
eastern Germany and western Russia, there is a period of fine 
anticyclonic weather known as the "Old Wives' Summer"; it 
resembles the "Indian Summer" of North America. The 
average duration is from 24th September to 4th October; it 
occurred in thirty-three of the fifty-two years 1 889-1 940. 
Similar, but shorter, more uncertain and progressively colder 
"late summers" may recur during October and November. 
Accounts of the European Monsoon and Old Wives' Summer 
are given by C. E. P. Brooks (1946a). 

In Scandinavia the eastward change of climate is much more 
rapid owing to the mountainous backbone of the peninsula, 
which shuts out the winds from the Atlantic. The climate of 
Norway is mild, rainy and stormy, with heavy snowfalls at high 
levels ; it resembles that of western Scotland, but is about 6° F. 
colder in winter at the same level. The climate of Sweden is 
much quieter and colder in winter and warmer in summer; it 
belongs rather to the east European type. The Gulf of Bothnia 
is largely frozen in winter (see p. 50) and this adds to the 



CLIMATES OF THE OLD WORLD IO5 

severity of the season. It is only in autumn that Sweden 
and western Finland take on the west European type of 
climate. 

eastern Europe and Siberia (Appendix I — Bulgaria, Sofia; 
Estonia, Tallin, Tartu; Finland, Helsinki, Oulu; Latvia, Riga; 
Lithuania, Kaunas; Poland, Danzig, Lwow, Warsaw; Roumania, 
Bucharest; U.S.S.R. (Europe), Archangel, Astrakhan, Lenin- 
grad, Lenkoran, Moscow, Odessa, Tiflis; (Asia), Barnaoul, 
Irkutsk, Markovo, Okhotsk, Sverdlovsk, Tashkent, Verkhoi- 
ansk, Vladivostok). 

The North European plain broadens eastwards and, apart 
from the Urals, which are not high enough to act as a real 
climatic divide, forms a monotonous plain extending from the 
southern mountains to the Arctic Ocean. In this great region 
all life is completely dominated by climate. There are three 
great zones extending from west to east. The mountains which 
form the southern boundary are bordered on the north by 
extensive steppes, becoming true desert in the Trans-Caspian 
region. Here winters are very cold, summers intensely hot. 
Where water supply is sufficient, or irrigation is practicable, 
this is rich agricultural country, but large areas of Asia support 
only a scattered nomad population with local aggregates in 
mining areas. Next comes a belt of forest from about latitude 
55 N. to the Arctic Circle. This is the most important part of 
the region economically, and the Trans-Siberian railway runs 
through it. The forests, in fact, supply most of the fuel for the 
locomotives. Finally comes the tundra, extending to the Arctic 
coast, a region of long, cold winters and short, dull summers, of 
little value and occupied only in small scattered settlements. 
The industries are mining and hunting or trapping. The Arctic 
climate is discussed on p. 165. 

The winters of Russia and Siberia are long and intensely cold, 
increasing in severity from the western borders to Verkhoyansk 
(67!° N., 133!° E.). Continuing the series on p. 104, but a 
little farther north, we have as the lowest temperatures on 
record (° F.) : Leningrad, —36; Archangel, —49; Barnaoul, 
—55; Eneseisk, —73; Verkhoyansk, —94; Okhotsk, —50. 
These figures sound impressive, but the cold is not as bad as it 
sounds. The breath freezes and falls in a white powder, but the 
air is dry and bracing, and during extreme cold there is never 
any wind. The hard, dry snow surface forms excellent going, 



106 CLIMATE IN EVERYDAY LIFE 

and winter is the favourite season for travel. One can move 
about freely on the level in calm air with a temperature of 
— 6o° F. (though the ascent of even a small hill renders breath- 
ing difficult), but a snowstorm at +5° F. is almost unbearable. 
The great drawbacks of the winter climate are the necessity for 
continuously wearing heavy furs, the impossibility of main- 
taining personal cleanliness, and the deadening monotony. 
Owing to the dryness of the air the winter climate is very 
healthy, lung complaints and all epidemics except smallpox 
being unknown. In spite of the apparently bracing quality of 
the weather, however, the cold saps the energy of the in- 
habitants. Western Russia and the bordering countries are 
fairly bracing and encourage mental and physical activity, but 
the climate becomes less favourable to the north, east and 
south. The difference is, however, one of degree rather than 
of kind, and the U.S.S.R. do not enjoy the variety of climate 
and resources which characterise the United States or the 
British Commonwealth. 

Although there are numerous light falls of snow the total 
amount is small, equivalent to less than a foot of undrifted snow. 
But the open country is swept by intense blizzards known as 
buran in south Russia and central Siberia and poorga in northern 
Siberia. These sweep up the dry snow in blinding sheets, and 
are very dangerous to travellers. They form drifts with inter- 
vening patches of bare soil through which the cold penetrates 
into the ground. 

The worst season comes at the end of winter — to call it 
"spring" would be a misnomer. Between March in the south- 
west and May in the north the ice breaks up in the rivers and 
drifts downstream, forming great ice-jams. Since the main 
rivers flow northward the ice breaks up first in the upper reaches 
and large areas in the valleys are flooded, while the melting of 
the snow and surface thawing cover the ground with sticky 
mud. At this season the country is almost impassable. Summer, 
though short, is hot and sunny (except near the Arctic coast, 
which has much fog at this season) and has a moderate rainfall. 
Since there is no danger of frost, vegetation flourishes and crops 
can be grown surprisingly far north. In late summer the 
climate of European Russia and western Siberia is pleasant, 
but eastern Siberia is damp and foggy, and this is the least 
healthy season. At this season the rivers are low and navigation 



CLIMATES OF THE OLD WORLD IO7 

is difficult. Winter begins rather suddenly in October; there is 
no "autumn." 

The east coast of Siberia has an unpleasant climate. Winter 
is not so cold as in the interior, but there are persistent strong, 
dry north-west winds, especially in the Amur valley. Summer 
is damp, cloudy and cool, with south-east winds which bring 
much fog and drizzle in the north and heavy rain in the Amur 
valley; the occasional west winds are accompanied by clear 
skies and swarms of mosquitoes. 

Northern Sakhalin has a most peculiar climate. Owing to 
the icy seas the coast is very cold, almost Arctic, and the damp, 
undrained valleys are occupied by peat-bogs and reindeer, yet 
at a moderate height in the interior the climate is almost sub- 
tropical. 

THE ALPS 

The mountainous region of the Alps and to a lesser extent the 
Carpathians and other ranges of central Europe are of interest 
chiefly for holiday resorts, winter sports, and sanatoria. They 
may be divided into mountain and valley climates. The effects 
peculiar to high altitudes are discussed in Chapter VII. Of the 
valleys all that can be said is that they are exceedingly diverse ; 
every valley has its own climate, and where a valley does not 
run north and south there is a further sharp distinction between 
the two sides of the same valley. Northward-facing slopes 
(ubacs) are especially unfavourable owing to the lack of sun- 
shine, and most settlements are situated on the southward- 
facing slopes (adrets). The most favoured spots are the 
"climatic oases," situated in wide valleys opening to the south 
and sheltered to the north and east by encircling mountains. 
Valleys opening to the north and east are intensely cold, and 
may be termed "little Siberias." 

A remarkable feature of the climate of high levels in the 
Alps is the large amount of sunshine in winter. At that season 
there is less cloud than in summer, and the higher resorts are 
generally above the level of the tops of the clouds, so that in 
spite of the shorter days there is almost as much sunshine as in 
summer. At Santis the average duration (hours per day) is: 
December, 4-3; April, 5-7; June, 4-8; August, 5-7. Owing to 
the thin atmosphere the sun's rays are but little weakened, and 
are reflected from the snow surface, so that in spite of the low 



108 CLIMATE IN EVERYDAY LIFE 

air temperature the body feels warm. The high-level resorts 
are especially rich in ultra-violet radiation, to which they owe 
much of their health-giving qualities. 

A characteristic of the Alps is the Fohn, a warm, dry wind 
which blows down the northern sides of the mountains whenever 
a southerly wind crosses the summits. It is best seen in the deep 
winding valleys which penetrate into the mountains; here it 
may blow on forty or fifty days a year, chiefly in autumn and 
winter. Because of the Fohn the winter temperature of Altdorf 
averages 3 F. higher than that of Zurich. The temperature 
may rise as much as 40 F. in a day. Snow melts rapidly and 
the thaw water causes sudden floods. The air is very dry and 
there is risk of fires ; in some villages smoking in the streets is 
forbidden during Fohn (W. G. Kendrew, 1930). Just before 
the onset there are rapid small oscillations of pressure which 
are said to cause nerve troubles. 

The rainfall of the Alpine districts is plentiful, and the 
water supply is maintained in summer by the melting of 
the glaciers and lower snowfields. This combined with 
the rugged topography and numerous mountain lakes has 
favoured the development of hydro-electricity, especially in 
Switzerland, where it is used to provide power not only 
for railways, street cars and factories, but in all the 
farms and in the home industries for which the country is 
famous. 

the mediterranean (Appendix I — Albania, Durazzo; Cyprus, 
Nicosia; France, Lyons, Marseilles, Nice, Ajaccio (Corsica); 
Gibraltar; Greece, Athens, Salonica, Gandia (Crete) ; Italy, 
Genoa, Milan, Palermo, Rome, Venice; Malta; Portugal, 
Lisbon; Spain, Barcelona, Cadiz, Madrid, Palma; Turkey, 
Ankara, Istanbul, Smyrna; Yugoslavia, Belgrade, Zagreb; 
Palestine, Haifa, Jersualem; Algeria, Algiers, Oran; Egypt, Alex- 
andria, Cairo; Libya, Benghazi, Tripoli; Morocco, Cape Spartel; 
Tunisia, Tunis). 

The region lying between the mountain ranges of the 
Pyrenees, Alps and Caucasus in the north, and the deserts of 
the Sahara and Arabia in the south, has mild, more or less 
rainy winters and long hot dry summers, a climate so charac- 
teristic as to be known as the "Mediterranean" type. Spain 
and Asia Minor form transition regions to the continental type ; 
the large land area and rugged topography make the climate 



CLIMATES OF THE OLD WORLD IOg 

of the interior very rigorous in winter and intensely hot in 
summer, and northern Spain has been described as "nine 
months winter and three months hell"; spring is the most 
pleasant season, but even that is subject to violent changes of 
temperature. The Mediterranean climate is seen at its best in 
southern France, northern and central Italy and the coast of 
Greece, where there is a fair amount of rain in all months and 
summer in not unpleasant, but the north coast of Africa, the 
lowlands of Asia Minor and Syria are nearly or quite rainless 
and very dusty from June to August or September. Over the 
whole region the rainfall is very variable, long droughts alter- 
nating with short periods of heavy rainfall. On the European 
coast heavy thundery rains amounting to 6 or 8 inches in a day 
are not uncommon, especially in spring, but on the African side 
heavy falls are much less frequent. Over the whole region there 
is little of the prolonged dull skies and steady rain of north-west 
Europe; the clouds build up rapidly in a blue sky and almost 
directly the rain stops the sun is shining again. On the Egyptian 
coast the rainfall is very scanty, averaging only 4 inches at 
Alexandria, 3 inches at Port Said and little more than an inch 
at Cairo ; the prosperity of Egypt is entirely due to the annual 
Nile flood. Irrigation is also necessary in parts of south-east 
Europe and south-west Asia, but in some of the backward 
countries it is rather crude. 

The Mediterranean is famous for its sunshine. Very few 
places outside the mountain districts have less than 2,000 hours 
a year, and the Egyptian coast exceeds 3,000, an average of 8 J 
hours a day. Even in December and January most parts 
receive more than 3 hours a day, the exceptions being the 
Balkan highlands and northern Italy; the latter is very misty. 
Spain and Portugal average 4! hours. In July and August 
the sun shines for 10 or more hours a day over the whole 
area. 

The winds are not so strong as in north-west Europe, but 
there are some important local winds. In the south of France 
occurs the well-known mistral, a violent north or north-west 
wind which is especially developed in the lower Rhone valley, 
where it is regarded with terror. It is very cold and dry, injuring 
delicate plants; gardens must be protected by high walls or 
thick cypress hedges. At Marseilles its average frequency is 1 10 
days a year, mostly in winter and spring, and it may continue 



110 CLIMATE IN EVERYDAY LIFE 

for a week, increasing in force in the afternoon and falling 
off again in the evening. It induces the irritable depression 
known as cafard. Very similar is the bora of the Adriatic, 
the Balkan Peninsula and the Black Sea near Novorossiisk. 
This is a dry, cold wind blowing in violent gusts from the north- 
east. It is most frequent in winter, when it may blow for weeks. 
In Gibraltar, a cool damp east wind, the levanter, is feared 
because it brings a feeling of lassitude. 

Opposite in character is the scirocco, an unusually warm 
southerly wind which prevails chiefly in Italy. As a rule it is 
damp and oppressive, bringing cloud and rain, and this is the 
characteristic wind of winter ; another form is a very hot, dry 
and dusty wind which occurs in Sicily and southern Italy. 
The temperature may rise to 95 F. at any hour, and the dust, 
which originates in the Sahara, is so thick as to colour the sky 
yellow and hide the sun. This wind is extremely drying and 
destructive to vegetation. There is no rain, or at most a few 
drops. This form of scirocco may occur in any month but is 
most frequent in spring. A similar wind, the leveche, occurs in 
Spain; in Madeira it is known as leste. The khamsin of Egypt is 
of the same character, remarkable for its extreme dryness ; at 
Cairo humidity has been known to fall to 2 per cent, and 
temperature to rise to 109 F. The khamsin is accompanied by 
sudden sandstorms in which the wind reaches gusts of 35 m.p.h. 
It is most frequent in March, April, and May. These hot, 
dust-bearing winds reach their greatest intensity in the dreaded 
simoom of the deserts of Algeria, Syria and Arabia (see p. 161). 
The coasts and islands of the western Mediterranean enjoy very 
favourable winters and are the region of winter health resorts 
par excellence. Spring is treacherous on the Mediterranean coasts 
of Europe and Asia, with sudden falls of temperature, especially 
in Palestine, which bring a feeling of greater cold than is shown 
by the thermometer. 

The long hot summer, which is prolonged into autumn, 
is unfavourable to exertion, and the Mediterranean is the 
home of the "siesta." Moreover, except in Spain and 
Portugal, Morocco and the western Riviera, there is malaria 
about. For these reasons the Mediterranean countries are 
not so advanced industrially as those farther north, where full 
activity can continue unchecked all day throughout the 
summer. 



CLIMATES OF THE OLD WORLD III 

Iraq, (Mesopotamia) and iran (persia) (Appendix I — Iraq, 
Baghdad, Basra, Mosul; Iran, Bushire, Teheran). 

The Mediterranean type of climate, with rainy winters and 
hot, dry summers, extends eastwards across south-west Asia, 
becoming more extreme. The winter rainfall is not heavy but 
is generally sufficient for pasture, and where irrigation is prac- 
ticable, as on the banks of the Tigris and Euphrates, rich crops 
can be grown, especially dates. The rivers are highest in 
spring, when the mountain snows are melting, and lowest in 
autumn, after the rainless summer. In the lowlands of the 
interior winter is a pleasant season, and the great heat of summer 
is made bearable by the dryness of the air. The Persian Gulf on 
the other hand is notorious for its damp sticky heat. The 
humidity at night is so great that everything becomes sodden, 
clothes, boots, books, go mouldy, and it is often noon before the 
sun dries out the moisture. Sleep is hard to achieve, and there 
is risk of heat stroke. 

The interior of Iran forms a high sub-tropical plateau, arid, 
intensely cold in winter, when the lakes and rivers freeze, no 
spring or autumn, and extremely hot and dry in summer, but 
with relatively cool and refreshing nights. At heights of 5,000 
to 6,000 feet the summers do not even feel hot, but a short 
exposure of the bare head to the sun will cause headache and 
probably sunstroke. Journeys are generally made by night or 
early morning. 

The Iranian province of Seistan is renowned for its winds. In 
winter true "blizzards" occur, one of which is said to have 
given a maximum wind speed of 120 m.p.h. In May and early 
June there is a respite, but then the "wind of 120 days" sets in, 
blowing from north-west with a speed which sometimes reaches 
70 m.p.h., and carrying huge quantities of dust. But at least 
the strong wind blows away the flies, which by this time of year 
are getting troublesome, and most of their attendant diseases. 
The wind also drives numerous windmills which raise the 
underground water into irrigation channels. 

Australia and new Zealand (Appendix I — Australia, Can- 
berra, Bourke, Sydney, Alice Springs, Brisbane, Normanton, 
Adelaide, Melbourne, Broome, Perth, Hobart; Mew Zealand, 
Auckland, Ghristchurch, Dunedin, Wellington) . 

These two dominions are conveniently included here because, 
although Australia extends far into low latitudes, the most 



112 CLIMATE IN EVERYDAY LIFE 

populous parts have a characteristic temperate climate. The 
northern coast of Australia, typified by Darwin, has a tropical 
monsoon climate, which extends down the east coast of Queens- 
land as far as Brisbane; this is described on p. 144. The basin- 
like interior is hot and dry and largely desert; the annual 
rainfall of Alice Springs for example is only 1 1 inches and is 
very irregular. The south coast from Perth to Adelaide has a 
"Mediterranean" type with hot, dry summers and equable 
rainy winters. In the interior flies are a nuisance even in winter 
and almost unbearable in summer. New South Wales and 
Victoria are definitely warm temperate; Sydney has a fairly 
heavy rainfall throughout the year, much of the summer fall 
coming in rather violent thunderstorms. There is a good deal 
of hail, and a summer seldom passes without a report from 
somewhere in the interior of hailstones as big as hens' eggs. 
The south and south-east coasts suffer from hot winds (brick- 
fielders) from the interior three or four times each summer; 
the dry heat injures vegetation. Sydney is protected from these 
by the Blue Mountains, but Melbourne has experienced maxima 
of over ioo° F. on six successive days. The climate of Tasmania 
closely resembles that of the maritime parts of southern England, 
allowing for the inversion of seasons ; Victoria is similar but a 
good deal warmer, and frost is almost unknown. Tasmania and 
Victoria have the most stimulating climate of Australia, but 
probably do not equal England in this respect. 

Australia is moderately stormy. Tropical cyclones known as 
" willy-willies" occur on the north-west coast between late 
November and late April, and have produced gust velocities 
exceeding 100 miles an hour. In the east also tropical cyclones 
approach the coast of Queensland in late summer and autumn, 
but their tracks are mostly offshore. When they strike the land 
they do a great deal of damage, both by the winds and by the 
floods which follow the very heavy rainfall. At Crohamhurst 
35 inches has been recorded in a day and 77 inches in four days. 
Tornadoes of the American type, but less severe, are occasionally 
met with in the north and east; for example, a tornado-like 
thunderstorm visited Sydney in January 1889. The best known 
storms are the "southerly bursters" which occur mainly in late 
spring and summer, and reach their greatest intensity on the 
coast of New South Wales. These are violent squalls in which 
the wind changes suddenly from north to south, with a great 



TEMPERATE CLIMATES OF THE OLD WORLD II3 

drop of temperature. Storms in Victoria may bring rainfalls of 
10 inches or more in a day in the mountains, and cause severe 
floods in the Darling River. 

Australia's chief trouble, however, is shortage and unreliability 
of rainfall, much of the centre and south having an annual total 
of less than 10 inches, while evaporation is very great, ex- 
ceeding 100 inches a year in the dry regions. Most of the rain 




Fig. 13. — Mean Annual Rainfall of Australia, inches. 

here falls in short heavy showers which do little good. The 
smaller rivers dry up in the dry seasons and the lakes are reduced 
to mud holes or salt pans, so that there is little chance of 
irrigation except where artesian wells yield a good supply. 
The artesian wells of eastern Australia are famous, and have 
done a great deal to aid the development of the drier regions. 
The supply of underground water is limited however, and is 
beginning to show signs of exhaustion. The total yield now is 
smaller, in spite of the greater number of wells, than when the 
water-bearing layer was first tapped. In the marginal areas 
h 



ii4 



CLIMATE IN EVERYDAY LIFE 



agriculture is very precarious. The following table shows the 
annual variation of rainfall at selected places (inches) : 





Lat. S. 


Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Queensland 
Thursday Is. . 
Cooktown 
Rockhampton . 


10 34 
15 28 
23 24 


17-4 
14-4 
77 


16-2 
13-8 
7-7 


13-8 
15-3 
4-3 


8-o 
8-9 
2-5 


17 
2-9 
i-6 


o-5 
2-0 
2-6 


04 

I'O 

i-8 


0'2 
I'2 

o-8 


01 
o-6 
i-3 


0-3 

I'O 

1-8 


i-4 
2'5 
2-4 


7-4 
6-6 

4-8 


New South 

Wales 
Sydney 


33 52 


3-6 


4-2 


4-9 


5-5 


5-o 


4-8 


4-8 


2-9 


2-9 


2-8 


2-8 


3-o 


Canberra 


35 20 


i-8 


17 


2-2 


i-6 


2-0 


2-1 


1-9 


2-1 


17 


2-1 


2-0 


2'I 


Victoria 
Melbourne 


37 49 


1-9 


17 


2-2 


2-3 


2-1 


2-1 


1-9 


1-9 


2-3 


27 


2-2 


2'3 


Tasmania 
Hobart . 


42 53 


1-9 


i-5 


17 


i-8 


1-9 


2-2 


2'2 


1-9 


2-1 


2-2 


2-5 


2-0 


Northern 

Territory 
Darwin 


12 28 


15-8 


12-9 


9-9 


4-2 


07 


0-2 


o-i 


01 


o-5 


2'2 


4-9 


io*4 


South Australia 
Alice Springs . 
Adelaide 


23 38 
34 56 


i-8 
07 


17 
07 


1-2 
I-O 


o-8 
17 


07 
27 


0-6 
3-1 


o-4 
2-6 


0-4 
2-5 


0-4 

2-1 


07 
17 


I-O 

I-I 


i-5 
i-o 


West Australia 
Broome . 
Carnarvon 
Perth . 
Albany . 


17 57 
24 54 
3i 57 
35 2 


6-2 

0-3 
0-3 

09 


6-i 
0-9 

o-5 
0-9 


3-8 
o-5 
07 
i-6 


1-4 
o-6 
i-6 
2-8 


o-6 

i-5 
4-9 
5-o 


I'O 

2-8 

6-9 
5-5 


0-2 
17 

6-5 
57 


0-2 
07 

57 
5-3 


o-i 
0-3 
3-3 
4-2 


o-o 
o-i 

2-1 

3-2 


0-9 
o-o 
o-8 
i-4 


37 
o-i 
o-6 

1-2 



The distribution of annual rainfall in Australia is shown in 

Fig. 13- 

The climate of Australia favours outdoor occupations 
(including cricket!), and as over most of the country except the 
north and east coasts the rainfall is too small and uncertain for 
large-scale agriculture, sheep-raising is the main industry. 
Except in the south-east the summers are too hot for sustained 
physical work by white labour. 

New Zealand extends over more southerly latitudes than 
Australia and, being narrower, is more maritime. The climate 
of North Island is warm temperate, equable, and rather rainy. 
The western side of the mountainous South Island is also equable 
and very rainy, resembling the wetter highlands of western 
England ; by contrast the eastern side is much drier, but the 
climate is nowhere extreme. As in Britain, summer is prolonged 
into autumn and winter into spring; the latter is an uncertain 
season, liable to cold snaps and late frosts. Climatic "accidents" 
are rare, however, and on the whole the climate of New 
Zealand is stimulating and very favourable for white people. 



CHAPTER IV 
TEMPERATE CLIMATES OF AMERICA 

IN this chapter we discuss the climates of the U.S.A. and the 
temperate parts of Alaska and Canada, southern Chile, 
Patagonia, and the Falkland Islands. The Arctic parts of 
Alaska and Canada are described in Chapter VII. 

THE U.S.A. AND THE TEMPERATE PARTS OF ALASKA AND CANADA 

Canada and the U.S.A. lie mainly in the zone of temperate 
west winds, but Canada and Alaska extend northwards into 
true Arctic regions, and in the Gulf of Mexico the U.S.A. reach 
the sub-tropical zone of east winds. North America resembles 
Eurasia in being a large land-mass with a highly continental 
climate in the interior and east, but in North America the main 
mountain ranges run north and south instead of east and west. 
In Eurasia the prevailing westerly winds carry the influence of 
the Atlantic Ocean far inland, at times across the whole of 
Europe and into Siberia. In North America the main mountain 
system, lying far to the west, forms an almost impenetrable 
barrier against the west winds from the Pacific. Instead, over 
most of the continent, north and south winds have free play. 
There is no permanent winter anticyclone like that which forms 
every year in Siberia; over most of the continent weather is 
governed by the passage of a series of depressions and anti- 
cyclones from west to east. In front of each depression the winds 
are southerly, and the influence of the warm Gulf of Mexico is 
carried far to the north. In the rear of the depressions the wind 
is north-westerly, and cold waves sweep as far south as the Gulf 
States and Florida, and even into Mexico and Central America. 
Stable climates are found only on the Pacific coast of California 
and in the Great Basin between the Rocky Mountains and the 
coast ranges of the south-western U.S.A. 

In the absence of a permanent winter anticyclone the interior 
does not suffer from the intense steady cold of Siberia. Instead, 
the whole eastern and central parts of North America have a 
regime of violent alternations of warmth and cold, including 

"5 



Il6 CLIMATE IN EVERYDAY LIFE 

the characteristic "cold waves" of winter and "heat waves" of 
summer. In the U.S.A. the official definition of a cold wave 
for forecasting purposes is a fall of 20° F. in twenty-four hours 
( 1 6-1 8° F. in the south and south-west) to a limiting tempera- 
ture which varies from o° F. in the northern interior to 32 ° F. 
in the south and south-west, but much greater falls than these 
are sometimes experienced (see p. 125). When the cold wave 
is accompanied by strong winds and the ground is covered by 
loose snow, "blizzards" result (p. 125). Except on the Atlantic 
and Pacific coasts there is a fairly regular snow cover for at least 
a month in winter down to about latitude 37 over most of the 
country. The low winter temperatures stop all outdoor work 
except lumbering (in which the snow cover facilitates the re- 
moval of timber) and the cutting of natural ice. 

In summer the hot moisture-laden winds from the Gulf of 
Mexico carry uncomfortable, sultry conditions far to the north. 
In the southern U.S.A. the damp heat of many summer after- 
noons makes factory work uneconomic. The principal manu- 
facturing region lies between this southern hot belt and the 
northern region, north of about 48 N., where transport is 
regularly blocked by deep snow and frozen waterways. 

The highly variable climate of most of North America has 
several interesting consequences. It is very stimulating, and 
probably accounts in large measure for the restless energy of 
the people of the northern part of the continent. Most of the 
theories about the effect of weather on work and business 
originated in the U.S.A., no doubt because of the variability 
of the weather. In the east of North America from the St. 
Lawrence to the Gulf of Mexico the rise of mean temperature 
from north to south is greater than over a comparable distance 
anywhere else in the world. Communication between north 
and south is easy, and the products of temperate and tropical 
regions are readily exchanged. There is also a great volume of 
holiday travel, south in winter, north in summer. 

As a whole North America is a sunny continent, the excep- 
tions being the Pacific slopes north of about 42 ° N. and the 
Great Lakes — St. Lawrence region. The rainfall comes in 
heavier showers on fewer days than in western Europe, the 
number of raindays being between 80 and 120 over most of the 
United States (Lake region 170, North Pacific coast 180). 
Thunderstorms are often violent, bringing heavy rain and 



TEMPERATE CLIMATES OF AMERICA 1 1 7 

squalls, which in their most intense form become tornadoes 
(p. 218). Thunderstorms are often accompanied by hail, large 
hailstones, the size of eggs or larger, doing great damage. At 
Cheyenne, Wyoming, destructive hailstorms are so frequent 
that glass-houses are protected by chicken netting. In the 
western plains and the deserts the thunderstorms are often 
"dry," the squall winds bringing dust-storms instead of rain. 
In such storms wire fencing, unless properly earthed at frequent 
intervals, is very dangerous because lightning, striking any part 
of it, is conducted along the wire and the whole fence becomes 
electrified. 

The winters vary greatly from year to year. On the Atlantic 
and Gulf coasts a winter month may be 8-io° F. warmer or 
colder than the long-period average. On the Pacific coast and 
in the south-west the range is little more than half as great. 

Annual "recurrences" like the stormy and anticyclonic 
periods of north- western Europe (see p. 102) do not seem to 
occur to the same extent as in that region, but this may be 
because they have not been studied in such detail. There is 
said to be a tendency for the last week of January to be rela- 
tively mild (the "January thaw"), followed by a return of cold 
conditions in the first week of February. Better known is the 
Indian Summer, the counterpart of the "Old Wives' Summer" of 
Europe. This is a period of calm, sunny days, dry, mild and 
hazy, which tends to follow the first severe frost and persist 
for several days. Like the Old Wives' Summer it is irregular 
in its occurrence; in some years there is no Indian Summer, in 
others there may be several such periods. 

North America is climatically "rich." The great variety 
of climates gives a corresponding variety of agricultural and 
forest products, but as similar climates often extend over large 
stretches of country, for example the "Great Plains," bulk 
production is facilitated. The mountains of the west, and the 
great eastward-flowing rivers such as the St. Lawrence system, 
provide abundant water-power (see p. 33), and where this is 
absent, on the plains, its place is often taken by wind-power. 
Finally, the variable stimulating climate of much of the country 
leads to a high level of production and also, probably, plays 
a part in the high rate of consumption which necessitates a 
high standard of living. Against this must be set the magnitude 
of the climatic catastrophes — blizzards, ice-storms, tornadoes, 



Il8 CLIMATE IN EVERYDAY LIFE 

hurricanes, great floods, droughts and dust-storms. These 
climatic "accidents" receive a good deal of notoriety, and the 
following pages may give the impression that the inhabitants of 
North America are always liable to be frozen, melted, blown 
away, washed away, dried up or choked by dust. That would 
be a gross exaggeration; North America is a large continent 
and climatic "accidents" are local; probably the majority of 
the inhabitants go from the cradle to the grave without suffering 
from any great climatic disaster. 

Climatically North America, excluding Mexico and the 
Arctic belt from northern Alaska to Labrador, may be divided 
into seven regions : — 

(i) Nova Scotia, New Brunswick and Newfoundland. 

(2) The large region, originally forested, extending from 
about ioo° W. long, to the Atlantic coast, excluding the most 
maritime parts of Canada, east of Quebec. 

(3) The Gulf States, including Florida, Louisiana, the 
southern halves of Georgia, Alabama and Mississippi, and south- 
eastern Texas. 

(4) The "Great Plains" region of small or deficient rainfall 
between about ioo° W. long, and the Rocky Mountains. 

(5) The northern Plateau from the interior of Alaska to 
about 42 ° N. 

(6) The southern Plateau and the Great Basin. 

(7) The Pacific slopes. 

NOVA SCOTIA, NEW BRUNSWICK, NEWFOUNDLAND (Appendix I — 

Charlottetown, P.E.I. ; Halifax, N.S.; St. John, N.B.; St. 
John's, Newfoundland). 

The easternmost provinces of Canada, and Newfoundland, 
have a rather inhospitable maritime climate with cold, damp 
winters and cool, foggy summers. The rainfall is heavy and is 
distributed fairly evenly through the year, with a maximum in 
late autumn and winter. The coasts of Newfoundland are often 
blocked by ice in spring (p. 49). The mean temperature varies 
from about 22 F. in January and February to 60-65 F. in 
July, with an annual mean of about 42 ° F. The climate is not 
suitable for agriculture; the chief industries are lumbering and 
fishing. 

The coasts are very stormy. In late summer and early 
autumn West Indian hurricanes, after travelling northwards 



TEMPERATE CLIMATES OF AMERICA Iig 

off the coast of the U.S.A., may strike Nova Scotia. The 
"cyclone" of August 1873 played havoc with the fishing fleets 
and is said to have wrecked 1,223 vessels, with a loss of over 200 
lives, but hurricanes of this intensity only occur at intervals of 
many years. In other seasons, especially winter, the depressions 
of middle latitudes tend to converge on this region from all 
parts of Canada and U.S.A.; though less intense, they are far 
more frequent than tropical hurricanes. 

eastern north America (Appendix I — Canada, Montreal, 
Ottawa, Quebec, Toronto, Winnipeg; U.S.A., Boston, Chicago, 
New York, Philadelphia, St. Louis). 

The extensive region between about 33 and 55 N. lat. and 
from ioo° W. long, to the Atlantic coast has an extreme climate 
with rapid changes of weather and a large annual range of 
temperature. The temperature rises steadily and rapidly from 
north to south; the following table gives a broad picture: — 

Table 15. — Mean temperatures, eastern North America. 





January 
°F. 


July 

°F. 


Year 
°F. 


Range 
°F. 


Canada, 45-53 N. 
U.S.A., 45-48 N. . 


8 
12 


68 
66 


40 
40 


60 
54 


40-45 N. . 
33-40° N. . 


23 
36 


73 
79 


50 
60 


50 
43 


For comparison, 
London, 5 1 £° N. 


41 


63 


50 


22 



The figures for U.S.A. are from R. de C. Ward (1925). 

The climate is continental; even the Atlantic coast does not 
enjoy a maritime climate, because the prevailing winds blow 
off the land. The Great Lakes have more effect in stabilising 
the climate than does the Atlantic, and the winter at places like 
Toronto is much less severe than farther west (e.g. Winnipeg). 
The promontories of the Lakes region have a very favourable 
climate; the Lakes prevent late frosts and these districts are 
famous for growing fruit and tobacco. 

The open plains give free play to the winds, which blow 
steadily though not strongly. The average wind speed is 8-10 
m.p.h. over much of the area, exceeding 12 m.p.h. in the Great 
Lakes region (Chicago is known as "the windy city "), the St. 
Lawrence valley and the more exposed parts of the Atlantic 
coast. 

Rainfall in the east is plentiful and well distributed through 



120 CLIMATE IN EVERYDAY LIFE 

the year, but it decreases westwards from 40-50 inches on the 
coast to about 20 inches in ioo°W., where it falls mainly in 
summer. North of about 37 N. the winter precipitation comes 
mainly in the form of snow, and a regular snow cover forms 
(Fig. 8). The depth of loose snow exceeds 8 feet south-east of 
the St. Lawrence and south of Lake Superior, and is 2 or 3 feet 
over all New England and the northern part of the region. 
Severe snowstorms occur, some of which have become historical 
because of the interruption of communications. Descriptions of 
some of these are given by C. F. Brooks (1935). On nth-i4th 
March 1 888 more than 3 feet of snow fell in three days and the 
gales piled up drifts 40 feet high. On 1 ith-i4th February 1899, 
44 inches fell in southern New Jersey. On i9th-20th February 
1934 Connecticut had 2 feet of wet, sticky snow, which was 
frozen hard on the streets and railway tracks by a succeeding 
cold wave. In the same winter snow fell to a depth of 8 feet in 
eastern Maine, and 15 feet deep on the south shore of the Gulf 
of St. Lawrence. 

"Ice storms" (glazed frosts, see p. 47) may be very severe in 
the northern part of the region. In November 1921 freezing 
rain fell for three days with a north-east wind in Massachusetts. 
Telephone wires had a coating of ice 2 inches thick and the 
supporting poles were snapped in hundreds. Communications 
and electric supply were interrupted for days. An even worse ice 
storm occurred in the Great Lakes region on 2 1st February 1922. 

The characteristic winter weather is an alternation of severe 
cold waves and moderate warm spells. The cold waves are 
sometimes accompanied by strong winds which raise the surface 
snow in blizzards (see p. 1 25) . Temperature may fall 30-40 F. 
in twenty-four hours. This kind of weather is invigorating to 
those in good health, but the extremes of damp cold and the 
artificially hot dry air of buildings induce respiratory diseases, 
which are especially prevalent in late winter and early spring. 
In eastern U.S.A. there are three or four severe cold waves in 
an average year. 

Spring is an uncertain season, periods of abnormal warmth 
alternating with cold spells and destructive frosts. In New York 
temperature in March has ranged from 8o° to 3 F., and in 
April from 91 ° to 12 F. In the St. Lawrence and lower Great 
Lakes the season is rather foggy. Spring is short, especially 
in the northern interior, where it is practically limited to April. 



TEMPERATE CLIMATES OF AMERICA 121 

The months of January to May bring the worst floods, 
especially in the valleys of the Ohio and Mississippi. The Ohio, 
because of its narrow valleys with steep gradients and the heavy 
rainfall on the Alleghenies, is especially liable to flooding. The 
floods result from heavy rain on frozen or waterlogged soil. 
The rain is due to warm, moist air from the Gulf of Mexico 
meeting cold air from the north-west ; melting snow rarely plays 
much part in causing floods. An exception was the record Pitts- 
burgh flood of March 1936, which was due to a combination 
of intense rain, the melting by a warm wind of a deep snow 
cover on the New England hills, and frozen ground which pre- 
vented the absorption of the water. The most disastrous flood 
was that at Johnstown, Pa., in May and June 1889, when 9-8 
inches of rain fell in thirty-one hours. Under the pressure of 
flood water a dam burst, and about 9,000 people were drowned. 
The worst occurrence of recent years was the Kentucky flood of 
January to February 1937, when the River Ohio rose 29 feet 
above normal flood level. The rainfall was not excessive on any 
one day, but was widespread and very persistent. A list of great 
floods in the U.S.A., and a discussion of their causes, is given 
by C. F. Brooks and A. H. Thiessen (1937). 

Summer is generally a quiet, hot season. Its chief charac- 
teristic, especially in the east and the Great Lakes region, is the 
combination of high temperature and high humidity, which is 
very enervating. Heat waves occur with a light southerly wind 
and clear skies, the days becoming steadily hotter. The high 
humidity keeps the diurnal range of temperature small, so that 
during heat waves the nights are especially unpleasant. Diar- 
rhoea is a frequent complaint, especially among children, and 
deaths sometimes occur from heat prostration. These heat 
waves are most frequent in the Mississippi Valley, but occur 
as far north as Montreal and Quebec. On the hottest days 
the sea breeze provides some relief within a few miles of the 
Atlantic coast, and to a lesser extent the lake breeze on 
the shores of the Great Lakes. The coast of Maine suffers 
from summer fogs which interfere with the summer holiday 
resorts. 

Summer is the rainy season over the whole interior, much 
of the rain coming in severe thunderstorms. These are often 
accompanied by squalls which may do considerable damage 
and are sometimes reported as "tornadoes." In spite of the 



122 CLIMATE IN EVERYDAY LIFE 

heavy rain the season is marked by abundant sunshine, the rain 
being intense and of correspondingly short duration. 

The western and central parts of the region sometimes 
experience severe droughts such as that which began in 1930 
and returned in 1934. The soil was dried out and blown away 
in great clouds of dust. Although the deficiency of rainfall was 
probably no greater than has occurred at intervals of thirty or 
forty years for many centuries, the increasing population and 
the withdrawal of ground water by pumping from wells and 
drainage of swamps, has accentuated the consequences. 

In late summer and autumn West Indian hurricanes (see 
p. 225) occasionally travel up the coast, bringing gales and 
heavy rain, which may result in disastrous floods in the eastern 
U.S.A. Examples were the floods of July 1916, due to heavy 
rain on the southern Appalachians, the floods in New England 
and New York in November 1927, when 8-77 inches of rain 
fell in a day in Vermont, and the New York flood of July 1935, 
with 9-0 inches in a day. In the 1927 flood there were 8-10 
feet of water in the business section of Montpelier, Vt., and at 
White River Junction the Connecticut rose 29 feet in twenty- 
four hours. As a rule summer floods are less serious than those 
of winter and spring because, although the rainfall is heavier, 
it is also more local, the ground is drier and there is a good deal 
of vegetation to retard the run-off. The West Indian hurri- 
canes also bring severe gales to the Atlantic coast, wind speeds 
reaching 75-90 m.p.h. The most disastrous of these swept the 
populous coasts of Long Island and southern New England 
on 21st September 1938. The damage to buildings, trees, 
power lines, highways and shipping is estimated by I. R. 
Tannehill (1944) as between 250 million and 330 million 
dollars; the loss of life was about 600. At Blue Hill Observatory, 
Massachusetts, the highest wind speed was 121 m.p.h. over 
five minutes, with a peak of 186 m.p.h. 

the gulf states (Appendix I — Charleston, Galveston, Key 
West, Miami, New Orleans). 

* The States of Florida and Louisiana, south-eastern Texas 
and the southern halves of Mississippi, Alabama and Georgia 
form a southward extension of the eastern province of the U.S.A. 
Owing to the influence of the warm Gulf of Mexico the tem- 
peratures are much higher, especially in winter, the weather 
changes are fewer and less violent, and the rainfall heavier. 



S TEMPERATE CLIMATES OF AMERICA 123 

R. de C. Ward (1925) places the northern boundary along the 
annual isotherm of 65 F. He gives as the average temperature 
over the whole area: January, 5i°F.; July, 82°F.; Year, 
70 F. ; Range, 3 1 ° F. The region is mainly agricultural, pro- 
ducing cotton and tobacco as well as fruit and early vegetables 
for northern markets. The main industry is tobacco processing. 
In spite of the general mildness of the winters cold waves 
sometimes penetrate as far as the Gulf Coast, bringing killing 
frosts even to Galveston in Texas, but southern Florida generally 
escapes^ Key West, off the southern tip of the peninsula, has a 
tropical climate. The lowest temperatures recorded at stations 
from north to south are: Wilmington, N. Gar. (34° N.) 5 F.; 
Charleston (33 N.), 7°F.; New Orleans (30 N.), 7°F.; 
. Jacksonville, Fla. fqo°N.). io°F.: Tampa, Fla.^ (28^.), 
i9°F.f Key "West (24J N.), 4i°F. The effect of the short 
stretch of ocean between the mainland of Florida and Key 
West on the minimum temperature is very notable. Snow falls 
occasionally as far south as northern and central Florida in the 
east and over most of Texas in the west, but very rarely lies 
so far south. A great snowstorm disorganised traffic in Texas 
in December 1929 when 2 feet of snow fell at Hillsboro (32 N.) 
and 2 inches even on the coast. Snow falls regularly in the* 
interior north of about 30 N. 

The spring floods of the Ohio and upper Mississippi some- 
times extend down the river to New Orleans. The most notable 
example was the flood of April 1927, when prolonged and 
violent rains over the whole Mississippi basin raised the level of 
the river to such a height that the levees were breached in 
many places. No less than 28,573 square miles of land were 
flooded, and the damage was estimated as 284 million dollars, 
but the loss of life was comparatively small, thanks to the flood 
warnings. In addition to thunderstorms, from ioth-i4th April, 
no fewer than eighteen tornadoes were reported. 

It is of interest that one of the earliest records of the Mississippi 
valley describes a great flood in March and April 1543, which 
held up de Soto's expedition. 

The summers are long, hot, humid and enervating. Summer 
is the rainy season, in which the bulk of the annual rainfall of 
50-60 inches falls, largely in severe thunderstorms. On the 
Texas coast the summer is hotter and drier than in Louisiana, 
and the rainfall maximum is delayed until autumn. 



124 CLIMATE IN EVERYDAY LIFE 

The wind speeds are generally light (about 8 m.p.h., 10 m.p.h. 
on the coast), but the whole region is subject to hurricanes in 
autumn which are much more severe than farther north. They 
move in from east or south-east in about 25 N. lat. and usually 
turn northward up the Atlantic coast, some distance offshore, 
to Nova Scotia. Some, however, penetrate the Gulf of Mexico 
and either strike the coast of Texas or turn north along the 
central valleys, breaking up over the land with torrential rains. 
The one which devastated Galveston on ist-i2th September 
1900 is said to be the worst storm on record in the United States. 
The wind speed was much above 100 m.p.h. (the anemometer 
was blown away), and raised the level of the sea by 15-20 feet, 
flooding the whole of the island on which Galveston is built. 
Nearly half the houses were destroyed, with property worth 
30 million dollars, and more than 6,000 people were killed. A 
hurricane struck Louisiana on 20th September 1909, the storm 
wave flooding New Orleans, again with great loss of life and 
property. Still more severe was the storm of 29th September 
1915, when the wind reached 140 m.p.h. and a large number 
of buildings in New Orleans and neighbouring towns were 
wrecked. In the famous Florida hurricane of i8th-i9th 
September 1926 the wind between Miami and Palm Beach 
reached 130 m.p.h. An eighteen-story skyscraper was twisted 
so badly that it had to be demolished. The damage to property 
exceeded 100 million dollars, 327 people were killed and more 
than 6,000 injured. Most of the loss of life was due to the 
breaking of a dam separating the small town of Moore Haven 
from Lake Okeechobee. In all these hurricanes the loss of life 
and of shipping would have been much greater but for the 
warnings issued by the U.S. Weather Bureau. 

v^JHE "GREAT PLAINS" BETWEEN 100° W. AND THE ROCKY 

mountains (Appendix I — Canada, Edmonton; U.S.A., Denver). 
This elevated region, rising westwards from about 2,000 feet 
to a general level of over 6,000 feet, has a very continental 
climate. The mean temperatures are about: — 

January July Year Range 

F. ° F. ° F. ° F. 

Canada, 49-54 N. . 8 65 40 57 

U.S.A., 42-49 N. 15 69 45 54 

3 o-42°N. 35 79 57 44 

The figures for U.S.A. are from R. de C. Ward (1925). 



TEMPERATE CLIMATES OF AMERICA 125 

The rainfall is almost everywhere less than 20 inches a year, 
so that without irrigation agriculture is very precarious. The 
whole region is open to the winds, and in winter cold waves, 
snowstorms and blizzards sweep over the country, sometimes 
penetrating as far as New Mexico and Texas. Temperature is 
very variable and may rise or fall 50 F. or more in twenty-four 
hours. In the blizzard of 12th January 1888 in the Dakotas, 
two or three hundred people, caught in the open, were unable 
to find their way to shelter and died of exposure ; thousands of 
cattle were lost. The wind speed exceeded 50 m.p.h., with 
temperatures falling from 30 F. to — 20 F. in five hours. The 
word "blizzard" comes from the German blitzartig, lightning- 
like, and aptly describes the suddenness of onset of these storms. 
A similar visitation occurred in North Dakota and Minnesota 
on I3th-i4th February 1923, and interrupted train services 
for over a week. South of 40 N. the winter snow cover is 
irregular in occurrence. 

The opposite of the blizzard is the warm, dry chinook, which 
blows from the south-west or west and descends the slopes of 
the Rocky Mountains. Under its influence temperature rises 
rapidly, sometimes 20-40 F. in a quarter of an hour. At 
Medicine Hat, Alberta, temperature has risen 70-80 F. in a 
few hours, and the range between the highest and lowest 
recorded temperatures exceeds ioo° F. in each month from 
January to April. March has experienced a maximum tem- 
perature of 84 F. and a minimum of — 38 F. At the onset of 
a chinook, before the cold air has all been blown away, the 
temperature often fluctuates wildly for some hours, until the 
warm wind sets in steadily. It often skips a belt of about 100 
miles wide at the foot of the mountains, which remains as a 
pocket of cold air. Under the influence of the chinook the snow 
cover disappears rapidly by melting and evaporation and the 
ground soon dries out. This enables the cattle to feed, and in 
fact without the chinook cattle raising would be very difficult. 
It also helps to keep the railways from being blocked by snow, 
but the dryness brings the risk of forest fires. 

Except in the far north, summer is intensely hot, maxima 
exceeding ioo° F. over almost the whole region, in spite of the 
general elevation. It is made up of long spells of settled fine 
weather and drought interrupted by occasional violent thunder- 
storms. The hot waves are more bearable than those of the east 



126 CLIMATE IN EVERYDAY LIFE 

because the air is very dry, but they are more damaging to crops. 
The worst feature is the "hot wind," the summer equivalent 
of the chinook, which descends from the mountains in local 
blasts of hot air a few miles wide, raising clouds of dust. "Hot 
winds" occur from June to September, most often in July and 
August. They cause great irritability and insomnia, and have 
been known to interrupt rail traffic by springing the rails. They 
are most severe in Colorado and Wyoming. 

The winds have an average speed of 10-14 m.p.h., and being 
steady are admirable for driving windmills. Over the open 
plains thousands of windmills are in use for pumping water 
for irrigation and for generating electricity. 

THE NORTH PLATEAU REGION, ALASKA TO IDAHO AND OREGON 

(Appendix I — Canada, Dawson). 

This region consists of a number of steep, narrow valleys 
between the Rocky Mountain Divide and the Coast Ranges. 
It is not of great economic importance. The interior of Alaska 
and north-western Canada have a continental climate with long 
severe winters and short, hot summers. The snow cover reaches 
a thickness of 3-8 feet, but melts quickly in spring. The ice in 
the Yukon breaks up in May. 

Rainfall in the valleys is light, but snowfall on the peaks is 
very heavy, especially in British Columbia, Washington and 
Oregon. Here the summers in the valleys are hot and dry, 
and irrigation produces large fruit crops. 

the south plateau (Nevada, Utah, Western Colorado, 
Arizona, Western New Mexico, interior of California) (Appendix 
I — Salt Lake City, Yuma). 

The wide, lofty plateau between the Rocky Mountains and 
the Sierra Nevada has an extreme climate. According to Ward 
the mean temperature over relatively low ground ranges from 
51 F. in January to 91 ° F. in July, with an annual mean of 
70 F., but owing to the great variations of elevation and 
exposure such figures have little meaning. Rainfall is generally 
between 5 and 15 inches a year, sometimes less, and it falls 
mainly in short, heavy showers in summer, when evaporation 
is intense. On the mountains the rainfall is heavier and supplies 
water for irrigation, producing large crops. Land which cannot 
be irrigated is desert or semi-desert and is valueless except for 
minerals. The winter snowfall on the higher slopes of the Sierra 
Nevada and Cascade Ranges amounts to 30-40 feet uncom- 



TEMPERATE CLIMATES OF AMERICA 127 

pacted, but it is drifted and packed by the wind and melts 
slowly, maintaining a good supply of water in the rivers. Snow 
sheds are necessary on the railways, and these are costly to 
build and maintain. In summer the occasional heavy rains 
cause landslides and washouts. 

The southern Plateau escapes the storms of higher latitudes 
and the winds are generally light. The dry, clean air is health- 
giving, and Colorado has a number of famous health resorts 
on the slopes above the valleys. The higher desert regions suffer 
from occasional strong dusty winds. The district includes the 
famous Death Valley in south-east California, a narrow de- 
pression 276 feet below sea-level, which has a mean July tem- 
perature of 101 F. and has recorded a maximum of 134 F., 
the second highest in the world. But in winter Death Valley is 
now a health resort. 

the coastal slopes of the pacific (Appendix I — Alaska, 
Sitka; Canada, Victoria, B.C.; U.S.A., Seattle, San Francisco, 
Los Angeles) . 

The narrow coastal strip which runs from Alaska to Cali- 
fornia is the only part of North America in which the climate 
resembles that of western Europe. The rainfall comes mainly 
in winter, and decreases from north to south. The winters are 
mild along the whole stretch of coast, the January mean tem- 
perature rising only from 32 ° F. at Sitka to 54 F. at Los 
Angeles. From Oregon northwards the weather is stormy and 
very rainy on the coast. Extraordinary wind speeds are re- 
corded on the bluffs of places like Tatoosh Island, where gales 
of 33 m.p.h. or more are experienced on ninety-six days from 
October to March, and the average wind speed for the whole 
year is nearly 20 m.p.h. The inland valleys are sheltered. North 
of San Francisco severe winter storms bring gales of 50-90 
m.p.h. In January 1921 a very heavy storm struck Washington 
and Oregon. At the mouth of the Columbia River the wind 
averaged 126 m.p.h. for five minutes and reached 150 m.p.h. 
in a gust. Enormous damage was done to standing timber. 

Snow falls regularly in winter north of 45 N., and occa- 
sionally as far south as Los Angeles. In Alaska it reaches 
depths of 4-12 feet, but in western U.S.A. it does not lie and 
is not important. 

Summer is a quiet settled season with moderately high 
temperatures. The July mean ranges only from 53 F. at Sitka 



128 



CLIMATE IN EVERYDAY LIFE 



to 70 F. at Los Angeles. The nights are cool and the afternoons 
warm. From Oregon northwards there is some cloud and rain, 
but California is the "land of sunshine" — and evaluates it in 
dollars and cents. Irrigation is necessary for raising crops, and 
the water for this is provided by the snows on the mountains. 
The regular sunshine favours the film industry and is also 
important for fruit drying. In spring and early summer 
California is subject to a very hot dry wind from the desert 
plateaus to the north or north-east, the Norther or Santa Ana, 
which resembles the "hot wind" of the Great Plains. It brings 
temperatures as high as 120 F. in June, and the heat and dry- 
ness damage fruit trees and increase the risk of fires. It also 
brings clouds of dust which are thick enough to impede traffic. 

California is generally rainless in summer, but at long in- 
tervals "cloudbursts" occur, and it is worth recording that at 
Campo, San Diego, 11-5 inches of rain fell in an hour on 12th 
August 1 89 1. This is the world's record for an hour's rainfall. 
On 25th August 1925, during a thunderstorm in the San 
Joaquin valley, a flash of lightning struck an oil reservoir and 
caused a great fire, which cost insurance companies more than 
a million dollars. 

The Pacific coast is very foggy in summer. The U.S. definition 
of "dense fog" is visibility less than 1,000 feet, and this occurs 
on forty or more days a year along the whole coast, mainly from 
July to September. San Francisco is famous for its fogs, and it 
is to escape them that the city has extended eastwards to less 
foggy and chilly sites in Berkeley and Oakland. 

The climate in San Francisco is most strange. It lies at the 
only gap in the Coast Range, separating the wide Sacramento 
and San Joaquin valleys from the Pacific, and in summer a very 
strong, steady wind pours through the gap, an intensified sea- 
breeze. This keeps the temperature down and brings in the 
sea fog. The mean and highest recorded temperatures are as 
follows : — 



Mean, ° F. 
Highest 



Jan. 



50 

78 



Feb. 



52 
80 



Mar. 



54 
86 



Apr. 



55 
89 



May 



57 
97 



June 
58 

IOO 



July 



58 
98 



Aug. 



59 
92 



Sept. 

61 
101 



Oct. 



60 
96 



Nov. 



56 
83 



Dec. 



51 

74 



As a local saying has it, overcoats are worn in summer and the 
lilies bloom in December. 



TEMPERATE CLIMATES OF AMERICA 129 

SOUTHERNMOST SOUTH AMERICA; FALKLAND ISLANDS 

South America extends from north of the Equator into high 
southern latitudes. Brazil and the northern countries are con- 
sidered under Tropical Climates (Chapter VI), Paraguay, 
Uruguay and most of the Argentine and Chile are sub-tropical 
(Chapter V). There remains only southern Chile, south of 
about 35 S., Patagonia, and the Falkland Islands. 

southern chile (Appendix I — Valdivia, Punta Arenas). 

The northern half of Chile has too little rainfall, the southern 
coastal regions too much. The annual totals increase from 5 
inches in 32 S. to 30 inches in 35 S., 40 in 37 , 60 in 38 and 
80 in 39 S. From Valdivia in 39 S. the rainfall exceeds 80 
inches over the coast and islands, and from 42 ° S. there is a 
narrow coastal strip with more than 100 inches a year, ex- 
ceeding 200 inches in places. The region is a vast morass, and 
for 900 miles the woods are so wet that it is impossible to set 
a fire for clearing without constant relighting. The prevailing 
winds are westerly and very stormy (the "roaring forties"), 
and these keep the temperatures moderate. 

Patagonia has a dry continental climate, with hot days but 
cool nights in summer, and cold winters. The only well- watered 
regions are in the valleys of the Cordilleras in the west, where 
there are a number of agricultural settlements. In the south the 
whole country is dominated by the blustery west wind. In spite 
(or because) of its severity the climate is healthy and invigorat- 
ing. 

the Falkland islands (Appendix I — Stanley) are cool, 
oceanic and so windy that the inhabitants are said to develop 
a characteristic walk. The western slopes are very inhospitable ; 
on the east there is some rough pasture, but the sunlessness and 
cold summer make the islands unsuitable for agriculture. The 
climate is very healthy. 



CHAPTER V 

SUB-TROPICAL CLIMATES WITH SUMMER 
RAINFALL 

NEAR the northern and southern tropics (23J N. and S.) 
the sun is nearly overhead for some hours at midday in 
summer, the weather is hot and there is a good deal of 
convection, so that summer is generally the rainy season. 
Where moisture-laden winds blow freely off the oceans the 
rainfall is considerable even on low ground; where mountains 
receive the full impact of these moist winds it is torrential. 
Winter by contrast is generally cool and dry. In Asia and to 
a less extent in West Africa and northern Australia there is a 
regular alternation of winds blowing from the oceans far into 
the interior of the continents in summer, and from the con- 
tinents to the oceans in winter. These are the monsoons which 
give rise to a special type of climate. Monsoons are most 
developed in southern and eastern Asia (India, Burma, Indo- 
China, China, Japan), not only because of the size of Asia, but 
also because of its rugged topography, which accentuates both 
the cold of winter and the heat of summer in the mountain- 
ringed basins. The West Indies, on the other hand, show the 
"pure" inter-tropical and sub- tropical climate almost un- 
disturbed by the effect of large land masses. South Africa, 
Madagascar, Mexico and Central America are intermediate 
in type. 

MONSOON CLIMATES OF ASIA 

The typical monsoon climates are found in southern and 
eastern Asia. There are three main types: (1) southern and 
south-eastern Asia; (2) China; (3) Japan. 

India, burma, indo-china (Appendix I — India, Allahabad, 
Bangalore, Bombay, Calcutta, Delhi, Karachi, Lahore, Madras, 
Peshawar, Simla; Burma, Mandalay, Port Blair, Rangoon; 
Indo-China, Phu-Lien, Saigon; Siam, Bangkok). 

This region, extending from the Arabian Sea to the Pacific 
Ocean, from 30 N. nearly to the Equator, and sheltered from 
north winds by almost continuous high ground, has a generally 

130 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 131 

warm climate and a well-developed alternation between north- 
east winds in the northern winter and south-west in the northern 
summer. The average temperature on low ground is above 75 
over the whole region and exceeds 8o° over all peninsular India 
except the west coast and all the more southerly parts of Farther 
India (Burma, Siam and Cambodia). In the southern and 
eastern parts of Farther India the air is moist, the relative 
humidity exceeding 80 per cent, in many coastal regions; other 
humid areas are the Assam Hills and the south-west coast of 
India, but over most of India the humidity is moderate and in 
Rajputana it is less than 50 per cent.; this region is com- 
paratively dry throughout the year. 

The distribution of rainfall shows a very wide range, from 
2 inches a year in north-west India to more than 400 inches 
near Cherrapunji in the Khasi Hills of Assam, where the 
average monthly distribution is (inches) : — 

Jan. Feb. Mar. April May June July Aug. Sept. Oct. Nov. Dec. 

o-5 3* 1 9-i VI 43'5 94*2 ioi-8 82-9 34-9 19-5 2-4 0-4 

The heaviest rainfall in twenty-four hours was 40-8 inches in 
June 1876. Another area of heavy rainfall is along the Western 
Ghats, between Bombay and Cochin and a third is the coast of 
Burma. In north-west India a large area, extending to the 
coast east of Karachi, is true desert except where it can be 
irrigated. In the interior of Burma, and in Siam and Indo- 
China, the rainfall is generally moderate. In Annam the 
rainfall is more evenly distributed through the year than in 
India. 

The number of raindays is not large because the rainfall is 
compressed into a short monsoon period. Exceptionally heavy 
falls are frequent; on seven occasions falls of more than 30 
inches in a day have been recorded in different parts of India. 

The alternation of monsoons dominates the life of India. 
Two seasons are distinguished, the dry season and the "rains," 
but the dry season is divided into the "cold weather" and the 
"hot weather." The cold weather lasts from the middle of 
December to the end of February, and on the low ground in 
the north is the pleasantest season of the year for Europeans, 
but Indians find the mornings too cold for comfort. Tem- 
perature is moderate, the whole region being below 8o° F. in 
January; in northern India the average is below 6o°, but 



132 CLIMATE IN EVERYDAY LIFE 

southern India is never really cool. The daily range is large, 
exceeding 25 over the greater part of the interior and reaching 
30 in many parts; hence though the days are hot the nights 
are cold. Temperatures below 45 ° have been recorded over the 
whole northern half of the region, and in north-west India 
frosts are common. Rainfall is very slight except in the north- 
west of India, where the cold-weather storms bring a small total, 
and on coasts exposed to north-easterly winds, such as the east 
coast of the Isthmus of Siam, which has its rainy season in 
November and December. 

The "hot weather" includes March, April and May, and is 
much less pleasant and more unhealthy than the winter. 
Temperature rises rapidly, and in May averages 93 ° F. in 
central India. The daily range is greater than in January, and 
very high maxima are recorded, above no° over north-west 
and central India and in the interior of Siam. In the Punjab 
the heat is aggravated by dust from the deserts to the west, and 
houses must be closed against it after sunrise, or sheltered by 
grass mats kept moist. 

In India the average duration of the south-west monsoon is : 
Bombay, 5th June to mid-October; Bengal, 15th June to late 
October; North-west Province, 25th June to 30th September; 
Punjab, 1 st July to mid-September. With the onset of the mon- 
soon rains the whole character of the climate changes. Except 
in north-west India temperature falls, and on the west coast of 
India and in southern Siam it is less than 8o° F. More im- 
portant is the decrease of the daily range, which is now less 
than 20°, so that while the days are cooler than in May the 
nights are not. Except in the north-west, the relative humidity 
is above 80 per cent, in most districts and exceeds 90 per cent, 
in parts of the Western Ghats and Burma. Cloudiness is very 
great, again excepting north-west India. In the rainiest areas 
rain falls nearly every day, but over most of the country the 
monsoon is broken by short periods of fine weather with rising 
temperature, which frequently end in thunderstorms. The 
rainy season is enervating, but not unhealthy; it is very destruc- 
tive of materials by corrosion and moulding. 

October and November in India are marked by lighter and 
more variable winds and decreasing temperature, cloudiness 
and rainfall in India. The excessive moisture on the ground and 
in the air, pools of water, and decaying vegetation under a hot 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 33 

sun make this season the least healthy and most liable to con- 
tagious diseases of the year, especially in the plains of northern 
India; damage by corrosion and moulding is even more rapid 
than during the rains. It is also the period during which severe 
cyclonic storms are most frequent in the Bay of Bengal. During 
the rainy months of June to September there are numerous 
moderate or weak storms, but severe cyclones are rare. The 
earlier transition period in May is a secondary maximum of 
severe cyclones; May, October and November account for 55 
per cent, of such storms. They mostly travel north-west or 
north, causing heavy rainfall on the coast, and hurricane winds 
which do much damage to property. If a storm comes at high 
tide a huge mass of water may be driven over the low ground, 
with much loss of life. These cyclones are the most destructive 
phenomena in India. As examples : in the terrible Backergunge 
cyclone of 1st November 1876 in Bengal a tidal wave struck the 
coast, about 100,000 people were drowned and all the crops 
were destroyed, causing famine and pestilence which were 
estimated to have cost a further 100,000 lives. Even worse was 
the cyclone of nth-i2th October 1737, when a storm- wave 
40 feet high swept up the Ganges and drowned about 300,000. 
Another storm- wave in 1864 devastated the same region. In 
the Octobers of 188 1 and 1924 typhoons caused great devasta- 
tion along the coast of Annam. 

At intervals of a few years the monsoon rains of India are 
deficient and the country suffers from severe famine, but good 
government and relief measures have largely overcome the 
effects of such seasons. Several irrigation projects have been 
carried out to relieve famines or to irrigate desert regions, and 
have brought excellent returns. Large areas of the lower Indus 
valley, formerly desert, have been irrigated by canals. The 
heavy rainfall of the seaward side of the Western Ghats mostly 
runs to waste in the Arabian Sea, while the Deccan to the east 
is dry, and in at least one case a river (the Periyar) has been 
diverted to flow eastwards through a tunnel in the mountains. 
Also, in order that relief measures may be taken in time, a 
great deal of effort has been put into attempts to forecast, 
several months ahead, whether the monsoon rainfall of the 
different parts of India will be above or below the average. 

The following data, extracted from the Climatological Atlas of 
India and the Indian Meteorological Memoirs, show the annual 



134 



CLIMATE IN EVERYDAY LIFE 



variation of the various climatic elements over the Indian land 
area. 

Table 16. — Climate of India 





Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Year 


Mean 




























Temperature ° F. 


67-5 


71-6 


79'2 


86-0 


88-7 


86-6 


83-5 


82-5 


82-5 


80.0 


73-6 


68-o 


79-0 


Daily Range ° F. . 


23-0 


23-5 


24-0 


23-0 


20-4 


157 


12-3 


I2-I 


14-0 


i8-6 


21-2 


22-3 


19-2 


Relative 




























Humidity % 


63 


58 


54 


52 


55 


67 


79 


8l 


79 


7o 


65 


64 


66 


Vapour 




























Pressure, mb. 


14-2 


147 


17-0 


19-5 


22-8 


26-6 


28-6 


28-5 


27-3 


22-6 


17-5 


14-8 


21*2 


Cloudiness, tenths . 


2-2 


2-3 


2-3 


2-6 


3-3 


5-6 


7-i 


6-q 


5-i 


3-i 


2-3 


2-2 


37 


Rainfall, inches 


o-5 


o-5 


07 


1-2 


3-i 


7-9 


II '2 


10-3 


7-o 


3-3 


i-3 


o-5 


47-5 


Raindays 


I 


i 


1 


2 


4 


9 


12 


12 


8 


4 


2 


I 


57 



The northernmost parts of India, especially the high-level 
valleys such as Kashmir, are almost temperate, and are prob- 
ably no less suitable for Europeans than are the Gulf States of 
the U.S.A., provided that reasonable sanitary precautions are 
taken and the head is kept covered against the sun in the hottest 
parts of the day. Bengal, the Ganges valley and farther south 
in India, and the central and southern parts of Burma and 
Indo-China, are too hot, and in many places too humid, for 
whites to maintain their health and efficiency over a number of 
years without periodic recuperation in a hill station or prefer- 
ably in a temperate climate. The loss of efficiency is probably 
due as much to tropical diseases, which can be avoided by care 
and sanitation, as to the direct effect of climate, except that the 
latter weakens the resistance to disease. But the necessary pre- 
cautions are expensive, so that the cost of living for whites is 
high. It is also difficult to raise healthy white children in south- 
east Asia, and children must be sent home to Europe at an 
early age. Apart from this, life in the more civilised parts 
of south-east Asia is pleasant enough, especially in the cold 
season. 

china, Mongolia, korea (Appendix I — Mongolia, Charbin, 
Urumtsi; Manchuria, Mukden; Korea, Jinsen (Chemulpo); China 
(North), Tsientsin; (Central), Shanghai, Nanking, Chungking; 
(South), Canton, Hong Kong). 

The climate of China is extreme, with cold dry winters and 
warm humid summers. The winter temperatures near the 
coast are the lowest for their latitudes anywhere in the world; 
in western China the winters are less cold because of the shelter 
afforded by the mountain ranges. The basin of Szechwan is 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 35 

especially favoured. Climatically the region falls into three 
zones : — 

(1) North China, Mongolia, Manchuria, Korea. 

(2) Yang-tse Valley. 

(3) South China. 

( 1 ) Most of the northern zone has very cold rainless winters, 
with temperature falling well below o° F. Strong dry north- 
westerly winds prevail, raising clouds of dust which are very 
troublesome in Peking, and sometimes interferes with naviga- 
tion on the coast. The Chinese take advantage of the wind by 
fixing sails to their wheelbarrows. Warm clothing is essential 
in spite of the bright sun. The rivers are frozen and a snow 
cover forms every year. The lowest recorded temperatures are 
(°F.): Charbin (46 N.), -36; Hsingking (Manchukuo, 
44 N.), -33; Si-wan-tse (4i°N.), -28; Taiyuan (38 N.), 
—21; Tsientsin (39 N.), —3; Tsinan (37°N.), +1. In Korea 
the cold is somewhat moderated by the neighbourhood of the 
sea, the lowest minima being —6° F. at Jinsen in the north- 
west and -f 7 F. at Fusan in the south-east. Eastern Korea also 
has some rain in winter, light in the north, moderate in the 
south. 

Summer lasts from May to September and is warm and 
humid, but not hot enough to be enervating; there is some rain 
from April or May to October, but more than half the annual 
total falls in July and August. In Manchuria in August the 
rain is almost continuous with frequent thunder, and the 
ground is flooded. The winds are south-easterly, mostly light 
or moderate. In the east the rainfall is sufficient for agriculture, 
but the western part of the zone is arid and much of it is 
desert. There is no real spring; the change from winter to 
summer is rapid. In some years the rainy season is delayed 
and hot dry west winds blow from March until June; these 
droughts do great damage to the growing crops. The river 
Hoang-Ho rises rapidly in summer and at intervals breaks its 
banks and over-flows the surrounding country, causing great 
disasters. 

In spite of the dependence of the country on agriculture, the 
great variability of rainfall from drought to flood, and the 
range of flow of the large rivers from winter to summer, there 
is only local irrigation in China, and nothing on the scale of the 



I36 CLIMATE IN EVERYDAY LIFE 

great enterprises of the Nile Valley. Local irrigation is mostly 
practised in the rice-growing districts of the south. The density 
of the population and the long courses of the rivers cause the 
water of the latter to be impure and unfit to drink without 
purifying or boiling. Since boiled water is rather unpalatable, 
this may be the reason why the Chinese have the universal 
habit of taking their fluid in the form of tea. 

(2) The Yang-tse valley has cold winters, with about three 
months below freezing, but much milder than those of the 
northern zone. Winter is the dry season, but there is some rain 
in all months ; the winds are lighter and, though most frequent 
from north, they are more changeable than in Manchuria and 
Mongolia. The sheltered basin of Szechwan has short winters 
with hardly any frost but almost continuous cloud, making 
them very depressing. The lowest minima are (from the coast 
inland, ° F.) : Shanghai, 10; Nanking, 8; Hankow, 13; Ichang, 
20; Chungking, 29. The summers are hot, with a mean tem- 
perature exceeding 70 F., and in the plains they are humid 
and enervating; July, August and September are the rainiest 
months. Szechwan is especially hot, the mean temperature of 
Chungking exceeding 8o° F. from mid-June to the end of 
August, while the maximum has reached in F. The moun- 
tain basins are subject to occasional severe thunderstorms, 
with heavy rain causing floods. The rainfall is much more 
reliable than in the northern region and severe droughts are 
unknown. 

(3) Southern China has generally mild dry winters with 
occasional cold spells when northerly winds from Mongolia 
extend to the south coast, bringing temperatures below 35 F., 
but the east coast iff" less subject to them. Thus the lowest 
recorded temperatures are 32 ° F. at Hong Kong, but 39 F. 
at Amoy. There is little rain from December to February, 
though no month is rainless and the winter is bright and sunny 
(the name Yunnan means "south of the clouds"). Frost and 
snow are very rare on low ground, but are frequent on the 
mountains which make up most of the region. Summer is hot 
and rather enervating in the plains, but healthy and pleasant 
in the mountains; the deep mountain valleys are, however, 
very unhealthy. The rainfall is very heavy in June, July and 
August. 

Between June and September the south-eastern coast is 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 37 

occasionally visited by typhoons (see p. 225), in which the wind 
speed may exceed 120 m.p.h. and the rainfall amount to 20 or 
more inches in a day. The typhoon which sank the De Witte 
in August 1 90 1 crossed Fukien and Chekiang. In August 1922 
a typhoon struck Swatow, and a great wave washed away the 
greater part of the city. In a typhoon which passed close to 
Hong Kong in August 1923 the anemometer recorded a gust 
of 127 m.p.h. These typhoons usually break up soon after 
crossing the coast, with thunderstorms and torrential rain. On 
19th July 1926, in such circumstances, 20-4 inches of rain fell 
at Hong Kong in nine hours (4 inches in one hour) ; the city 
was flooded and considerably damaged. In another typhoon 
the centre of which passed over Hong Kong on 2nd September 
1927, the anemograph showed a maximum velocity of 167 
m.p.h., but the rainfall was only 5-9 inches. A good deal 
of damage was done and the typhoon wave caused many 
deaths. 

japan, Formosa (Appendix I — Japan, Nagasaki, Naha, 
Nemuro, Otomari, Tokyo; Formosa, Taihoku). 

The northern part of Japan (Hokkaido and northern Honshu) 
has severe winters. The temperature is not so low as on the 
opposite mainland, but the western sides and the mountains 
have very heavy snowfall, depths of 20 feet or more burying the 
mountain villages and making all outdoor work impossible. 
The east coast is clear and dry. Southward the winters become 
much warmer, and the islands of Shikoku and Kyushu are 
almost sub-tropical, but even Formosa experiences occasional 
cold days. The lowest recorded temperatures are (° F.) : 
Otomari (Sakhalin, 46J N.), —27; Nemuro (43J N.), —9; 
Hakodate (41 N.), +2; Niigata (38 N.), 15; Tokyo (35J N.), 
17; Nagasaki (32J N.), 22; Taihoku (Formosa, 25 N.), 32. 
The mildest winters are in the Liu-Kiu Islands, the lowest 
temperature at Naha (26 N.) being 41 ° F. 

Summer is warm and humid over the whole region; in 
the north and at high levels it is a pleasant, invigorating 
season, but over most of Japan and in Formosa it is hot and 
enervating. 

On the western side of Japan and on the north-western tip of 
Formosa (Keelung) the rainfall is heavy throughout the year, 
with a maximum in winter where the prevailing north-west 
wind blows onshore. The eastern and southern coasts have 



i 3 8 



CLIMATE IN EVERYDAY LIFE 



only moderate rains in winter and heavy rains in summer; as 
shown by the following monthly averages in inches: — 



Niigata (West Coast) 
Yokohama (East Coast) 
Nagasaki (South Coast) 



Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


7-6 
2-8 
3-0 


5-o 

3-i 
3-3 


4-3 

5*2 


4-2 
5-5 
7'S 


3-6 
6-o 
6-7 


5-2 
7-i 

12-8 


6-3 
6-7 
9-4 


4'9 
8-7 
6-8 


7-6 
io-3 
8-6 


6-i 

8-5 
4-8 


7-3 
4*1 
3-5 



Dec. 
9-3 
3'4 



The heaviest rain in the east comes in the autumn and is partly 
associated with the passage of typhoons ; there is a secondary 
maximum in early summer. On the south coast and in the Liu- 
Kiu Islands the wettest period is June and early July, with a 
secondary maximum in September. The rainy period of June 
and July is known as the Bai-u or "Plum rains"; it is very im- 
portant for the rice crop, but the weather is most unpleasant, with 
very humid air, rain nearly every day and no sun. During this 
period leather and cloth go mouldy and stores deteriorate rapidly. 

Typhoons may traverse Formosa and the Japanese Islands in 
almost any month, but are only a serious threat from August to 
October, especially in September. The winds exceed gale force, 
but are not so strong as the typhoon winds of South China ; on 
the other hand they bring torrential rain. In southern Japan 
a fall of 35*5 inches has been recorded in twenty-four hours, 
of which 9*5 inches fell in two hours. In northern Formosa 
a fall of 40 inches has been recorded in twenty-four hours. 

In spite of these disadvantages the climate of Japan is not 
unfavourable to Europeans, especially in the north, which is the 
most bracing. Even in the south the enervating effect of the 
hot moist summers is largely counterbalanced by the cold 
winters. Judged by the industry and productiveness of the 
Japanese the climate seems to be healthy enough to favour a 
considerable output of energy. 



MONSOON AND SUMMER RAINFALL CLIMATES OF AFRICA 

The monsoons of Africa are less well developed than those 
of Asia, but in the west and south there is a clear distinction 
between the dry and rainy seasons. Owing to lower latitudes 
and smaller land mass the dry season is not cold. 

west Africa (Appendix I — French West Africa, Dakar, 
Niamey; Gambia, Bathurst; Sierra Leone, Freetown; Gold Coast, 
Accra; Nigeria, Kaduna, Lagos; Cameroons, Duala). 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 



139 



The monsoon area of West Africa extends from about 14 - 
4 N. The coastal regions are hot, moist, enervating and un- 
healthy; the mean annual temperature exceeds 8o° F. from 
Sierra Leone to Lagos. The more elevated and drier regions 
of the interior are healthier and more invigorating. The health 
of whites has improved in recent years, however, even on the 
coast, owing to improvements in habits and sanitation, and West 
Africa no longer deserves the title of "The White Man's Grave." 

North of about 8° N. the year is divided into two seasons, the 
cool dry season from about November to April, with dust-laden 
east or north-east winds, and the warm rainy season from about 
May to October with south-west winds; in the north (e.g. 
Bathurst) the rainy season is shorter and more definite and the 
dry season is rainless. South of 7 N. the dry season is less 
definite ; the rainy season is longer and is split into two parts by 
a short spell of less rain in August, which includes some bright 
clear days. Here the hottest month is March, before the begin- 
ning of the rains. South of about 6° N., however, this short 
dry season tends to disappear again, and Duala in 4 N. has 
only a single maximum in July. Between 4 N. and the Equator 
there is an almost complete reversal, July being the driest month. 

The rainfall is generally heavy on the coast, the wettest 
stretches being from io° N. to Gape Palmas and the north-east 
corner of the Gulf of Guinea ; on the Cameroon Mountains the 
rainfall of 369 inches is almost as heavy as anywhere else in the 
world. The Gold Coast is exceptionally dry for its situation. The 
amounts decrease inland, rapidly from south to north, more slowly 
from west to east, the country gradually becoming more arid until 
it passes into the desert of the Sahara. North of about 15 N. 
the average is less than 20 inches a year, which is insufficient. 
The following table (inches) illustrates the peculiarities : — 



Table 17. — Rainfall of West Africa 





Lat. ° N. 


Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Dakar . 


14* 


_ 


_ 


— 


— 





07 


3-5 


q-6 


5-4 


i-6 


o-i 


o-3 


Bathurst 


I3i 


— 


— 


— 


— 


0-2 


3-o 


10-9 


ig-6 


io-o 


37 


0-2 


0-1 


Freetown 


8* 


0-3 


0*2 


i-i 


2-8 


8-6 


18-2 


35-5 


32-7 


25-9 


io-8 


V6 


I-O 


Porto Novo 


6* 


O-Q 


1-2 


37 


4*9 


8-2 


12-7 


4-6 


1-2 


4-4 


7-1 


3'4 


o-6 


Lagos . 


6* 


I-I 


i-8 


4-0 


5-9 


io-6 


iX-i 


n-o 


2-5 


5-5 


8-1 


2-7 


I'O 


Accra 


5i 


o-6 


1-2 


2'0 


3'4 


5-3 


7-0 


i-6 


0'6 


i-3 


2-3 


i-4 


0-9 


Calabar . 


5 


21 


2-7 


6-4 


7-9 


n -9 


15-7 


i6-q 


16-2 


16-3 


I2-S 


7-5 


21 


Duala . 


4 


i-8 


37 


8-o 


9-1 


n-8 


21-2 


29-2 


27-3 


20-9 


l6'9 


6-i 


2-5 


St. Thomas Is. 





4-i 


4-3 


7-0 


5-o 


4-7 


o-5 




04 


09 


4-3 


5'7 


3-6 


Bolobo . 


2°S. 


5-o 


7-0 


4 b 


7--' 


5-6 


0-4 


~ 


27 


3-8 


6-5 


9-0 


IO-2 



140 



CLIMATE IN EVERYDAY LIFE 



The dry season is generally healthy. A cool dry east or north- 
east wind blows from the interior, known as the harmattan or 
locally as "the doctor," which in January extends to the coast 
at Lagos and Accra. It carries large quantities of fine dust 
which penetrates all crevices. On the coast there are strong 
land and sea breezes. In April and May the West African 
"tornadoes" occur, but these are not true tornadoes, merely 
thunderstorm squalls which occasionally unroof buildings or 
uproot trees. During the rainy season the winds are light; the 
end of this season is the most unhealthy time of year. 

north-east Africa (Appendix I — Egypt, Alexandria, Cairo; 
Soudan, Khartoum, Mongalla; Abyssinia, Addis Ababa; Red Sea, 
Kamaran Islands). 

Climatically Egypt forms part of the Sahara-Arabia desert, 
but its whole life is based on the annual Nile flood, which 
results from the monsoon rainfall over Abyssinia. Between 
Cairo and Atbara (i7°45 / N.) the annual rainfall is less than 
an inch and several years may pass without a shower. The sky 
is almost cloudless, the air is very dry and dusty, and the day 
temperatures are very high. The highest temperatures to be 
expected on the hottest day of the average month (i.e. the mean 
monthly maxima), the highest recorded and the average 
relative humidity, are as follows : — 



Table 18. — Climate of North-east Africa 





Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Cairo 


























Mean Max. ° F. 


74 


81 


9i 


100 


104 


106 


103 


101 


97 


96 


8q 


79 


Abs. Max. ° F. 


80 


92 


99 


109 


in 


109 


108 


106 


106 


100 


Q7 


82 


Rel. Hum. % 


59 


5i 


46 


4i 


38 


41 


47 


5i 


54 


53 


56 


59 


Khartoum 


























Mean Max. ° F. 


99 


105 


109 


112 


114 


112 


109 


105 


108 


108 


103 


99 


Abs. Max. ° F. 


103 


109 


112 


H.5 


116 


"5 


117 


109 


in 


109 


106 


104 


Rel. Hum. % 


26 


20 


14 


13 


18 


28 


43 


52 


42 


29 


25 


27 



With the highest temperatures the humidity may fall to 
2 per cent, and evaporation is very rapid. The nights are rela- 
tively cool, and the climate is not unhealthy. For the effect of 
the high temperatures see Chapters VII and VIII. On the 
Red Sea coast the days are almost as hot, but the air is moister 
and the nights less cool, making the climate very unpleasant. 

The worst feature of Egypt is the khamsin, a hot dry southerly 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL I4I 

wind which carries great quantities of dust and is sometimes 
accompanied by sand- and dust-storms. The khamsin is most 
frequent in spring. In the Soudan strong squally dust-bearing 
winds are known as haboobs. At Khartoum more than twenty 
haboobs occur each year between January and October; they 
are most frequent in June and July. They occur most often 
between 2 p.m. and 10 p.m. and last about three hours. 

The rainfall increases rapidly south of Khartoum, and the 
rainy season lengthens. At Khartoum (i5i°N.) 5-2 inches 
fall, almost all in July, August and September. At Malakal 
(9J N.) the total is 34-8 inches, from late April to early Novem- 
ber. The heaviest rain, however, falls on the mountain plateau 
of Abyssinia during the south-west monsoon, and averages 
nearly 50 inches a year. About half the annual total comes in 
July, August and September, but there is a little rain even in 
winter. This heavy fall runs off into the Blue Nile, and though 
about half of the water is lost by evaporation in the Sudd 
swamps, the remainder causes the annual Nile flood. From 
Khartoum northwards practically all cultivation depends on 
irrigation from the river or from irrigation canals fed by the 
river. The flood is very variable from year to year and dams 
have been built at intervals to store the water from good years 
and to regulate the flow. A comprehensive plan for further 
works, involving also the Soudan, which will double the irrigable 
area and also provide a supply of hydro-electricity, has been 
prepared for the Ministry of Public Works, Egypt, by H. E. 
Hurst (1946) and his colleagues, and is now being put into 
operation. 

Madagascar (Appendix I — Tamatave, Tananarive). 

Three climatic zones may be distinguished in Madagascar, 
the east coast, the central highlands and the west coast. The 
east coast (table for Tamatave) has a moderate temperature, 
with rain throughout the year, the only approach to a dry 
season occurring in September to November. The winds blow 
off the sea throughout the year, keeping the climate fresh and 
healthy. The central highlands (table for Tananarive) have 
also a moderate temperature, and in spite of the elevation the 
rainfall is not large. There is a well-marked winter dry season 
from April to October, but the humidity remains high. A 
peculiarity of the first part of the dry season is a kind of Scotch 
mist, very cold and wetting, but giving only small amounts of 



142 CLIMATE IN EVERYDAY LIFE 

rainfall. Snow is unknown, but frost is occasionally seen. During 
the rainy season heavy westerly squalls occur with thunder and 
violent rain. Temperatures do not exceed 95 in the highlands. 

The west coast is much hotter and drier than the east, the 
average annual temperature being 79 at Mojanga (16 S.). 
The rainfall decreases from north to south, averaging 62 inches 
at Mojanga, but only 14 inches at Nossi-Be (23 J° S.). The 
south-west coast is almost a desert, but is healthy. 

The cyclones of the Southern Indian Ocean occasionally cross 
the island, especially the northern half; they are limited to 
December to April and are most frequent in February. 

south Africa (Appendix I — Rhodesia, Salisbury; Union of 
South Africa, Gape Town, Durban, East London, Johannesburg, 
Kimberley; South-west Africa, Walvis Bay, Windhuk). 

Almost the whole of the continent of Africa south of io° S. 
is occupied by an extensive plateau, between 3,000 and 6,000 
feet high, rising in the Drakensberg to nearly 12,000 feet. The 
low ground forms only a very narrow fringe round the coast 
in the west and south, but broadens out in the east. The result 
of this topography is a great uniformity of climate over a large 
area, while the temperature is low for the latitude and resembles 
that of western Europe. Since the highest elevations are rather 
near the coast, rainfall over the interior is generally deficient. 
On the edge of the plateau where the air descends rapidly to the 
coastal plain, "hot winds" occur, similar to the Fohn and 
Chinook. These berg winds are most frequent on the south 
coast, where they blow on twenty to thirty days a year, mostly 
in winter, when they cause very high temperatures, often 
exceeding those of summer. When they blow for two or three 
days they are very oppressive. On 22nd January 1923 a tem- 
perature of 1 1 8° F. was recorded at Dunbrody in the south of 
the Cape Province during a berg wind ; mealies and other crops 
were destroyed and some cattle, ostriches, fowls and bees died. 

Dust-storms are frequent, especially from August to Decem- 
ber. The fine dust is raised by a strong, squally wind and is 
very penetrating, but the dust-storms rarely last long and are 
usually followed by rain. 

The climate is extreme; maximum temperatures exceeding 
ioo° F. have been recorded over most of the area, except at 
high-level towns such as Johannesburg, and all over the plateau 
95 is exceeded in most years. The minima are correspondingly 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 43 

low in the interior, and temperatures below 32 ° F. have occurred 
as far north as Salisbury. The coasts are less extreme and even 
the south coast is mostly free from frost. Relative humidity is 
moderate on the south and east coasts, but low in the interior, 
averaging only 54 per cent, at Kimberley. In spite of its desert 
character the west coast has a cold, damp and very foggy 
climate with, however, very little rain. 

The annual variation of rainfall shows a maximum in sum- 
mer and a minimum in winter over most of the area. Winter 
is very dry in Northern Rhodesia and Nyasaland, and in the 
interior of South Africa, especially Matabeleland, Damaraland 
and Great Namaqualand, and to a less extent in the Transvaal. 
The provinces of the east coast, Mozambique, Lourenco Marques, 
Zululand and Natal, have an appreciable winter rainfall. Here, 
in addition to the usual summer thunderstorms, rain is brought 
by gales from the sea, which may come at any season. In Natal 
these are known as "three-day rains" because they generally 
last two or three days. In summer cyclones from the Indian 
Ocean sometimes approach the coast of Natal and bring heavy 
rain. The following table of monthly rainfall in inches sum- 
marises the annual variation : — 



Table 19. — Rainfall of South Africa 





Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Mozambique and Natal 
Transvaal 

South Coast . ... 
South West Cape Province 
and Cape Peninsula 


5-8 
6-i 
i-8 


5-o 
5-3 

2-1 

07 


4-8 
2-4 
1*2 


2-8 
i-5 
2-3 

2-8 


i-6 
0-4 
2-3 

4-6 


0-9 
o-i 
1-9 

5-2 


o-6 
o-i 
i-6 

4-2 


i-i 
0-3 

2-2 
4-2 


i-4 
o-8 
2-6 

3-o 


2-7 

2'0 

2-7 
2-6 


3-9 
3-9 
23 

i'3 


4-7 
4-8 
2-3 

1-2 



Over the interior a large part of the rain falls in severe 
thunderstorms which occur on more than thirty days a year 
over most of the Transvaal, Basutoland and eastern Cape 
Province. Hail is remarkably frequent, occurring on 105 days 
a year in the Cape Province and Transvaal, mostly in summer. 
In the most severe storms some of the hailstones may be as big 
as cricket balls and weigh up to ij lb.; they kill sheep and cows 
and pierce corrugated iron roofs like paper. Fortunately tiiey 
are not usually accompanied by wind, though one violent 
hailstorm was preceded by a true tornado of the American 
type. Some heavy falls of rain have been recorded in thunder- 
storms, including one of 16-5 inches at Swellendam. 



144 CLIMATE IN EVERYDAY LIFE 

In the west of South Africa thunder is rare. The west coastal 
region of South-west Africa is true desert and uninhabited. 

South Africa is noted for its sunshine. This and the dry air, 
especially of the central and upper Karoo, make it very suitable 
for sufferers from tuberculosis and phthisis. In the Orange Free 
State and Transvaal cold wet weather is unknown, but the dust 
is liable to cause ophthalmia. 

northern Australia (Appendix I — Darwin, Thursday 
Islands). 

The northern coast of Australia lies well within the tropics, 
and has a typical monsoon climate. The north-west monsoon 
blows from December to March; in West Australia it becomes 
south-westerly and is cooler and drier. The south-east trade 
wind blows for the rest of the year. Owing to the warm seas to 
the north the climate is warm throughout the year, and is 
especially oppressive in April. The dry season from May to 
October has only a few showers; the rainy season comes in 
almost suddenly in December and torrential rain falls almost 
every day for three or four months. Inland the rainfall de- 
creases rapidly, especially in the west, and soon passes into the 
desert. For the monthly and annual rainfall see p. 114 and 

Fig. 13. 

Mexico and central America (Appendix I — Mexico, Mazat- 
lan, Mexico City, Salina Cruz, Vera Cruz; British Honduras, 
Belize; Costa Rica, San Jose; Guatemala, Guatemala; Panama, 
Balbas Heights, Cristobal ; San Salvador, San Salvador) . 

Mexico is a land of varied climates, from the hot, steamy 
southern shore of the Gulf of Mexico to the north-western 
deserts. A large part of the country consists of a lofty plateau 
over 6,000 feet above sea level. The inhabitants divide the 
country into three zones, the tierra caliente (hot lands) up to 
2,000 feet, the tierra templada or moderate zone from 2,000 to 
6,000 feet, and the tierra fria (cool land) above 6,000 feet, but 
even the latter has, over most of its extent, a hot summer 
climate. At Mexico City, 7,500 feet up, the mean temperature 
exceeds 6o° F. from March to September. In addition, there 
is a marked difference between the rainy eastern and dry 
western sides. The rainfall varies from over 100 inches on the 
slopes facing eastwards above the southern parts of the Gulf of 
Mexico, around Vera Cruz, to less than 20 inches over most of 
the interior plateau and less than 10 inches in Sonora and Lower 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 45 

California. On the southern Pacific coasts it increases again, 
amounting to 40 inches a year at Salina Cruz in Oaxaca. On 
the plateau water is scarce and has to be drawn from wells 
which may be as much as 100 feet deep. 

Over most of Mexico there are three seasons, the "cold 
weather" from November or December to February, the "hot 
weather" from March to May, the two together forming the 
dry season or verano, and the rainy season from June to October. 
In the south the rain is less heavy in July and August than in 
June and September, forming the lesser dry season or veranillo. 
On the Atlantic side this is barely perceptible, but on the 
Pacific side it brings a real break, with dry weather and a rise 
of temperature. At Salina Cruz the monthly rainfalls and 
daily maximum temperatures are as follows : — 



Dec. 



Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


0-2 
83 


84 


85 


0-3 
87 


3-0 

88 


14-6 
85 


2-4 

88 


3-8 

89 


10-2 

86 


5-4 
85 


0-3 

85 



Rainfall, inches . .0-2 — — 0-3 3-0 14-6 2-4 3-8 10-2 5-4 0-3 o-i 
Mean daily max. temp. °F. 83 84 85 87 88 85 88 89 86 85 85 84 



The "cold weather" is dry and sunny over most of the 
country. The prevailing winds blow from east or ENE. ; they 
are steady and stable and bring rain only on high ground 
facing the Gulf of Mexico. They are interrupted from time to 
time, especially in January and February, by cool northerly 
winds or JVortes, the continuation of the Northers of the United 
States. These are strong and stormy on the southern Gulf 
Coast, and bring persistent fine rains. On the western shores of 
the Gulf they are stormy and distinctly cold, and in the lee 
of the mountains they are cold and dry, and bring frosts ; some- 
times they descend the Pacific slopes as the hot dry dusty 
papagayo. The Gulf of Tehuantepec suffers in winter from the 
Tehuantepecer, a violent north wind which is a continuation 
of the cold waves from North America, blowing through 
the 70-mile wide gap in the Cordilleras in the Isthmus of 
Tehuantepec. 

In March, April and May the wind becomes more south- 
easterly and cloudiness begins to increase, but is not enough to 
balance the increasing heat of the sun ; the days are very hot, 
but the nights are cool. On the humid Atlantic lowlands whites 
cannot work in the open at noon and in the afternoon. The 
day temperatures at this season are even higher on the Pacific 



I46 CLIMATE IN EVERYDAY LIFE 

coast, but the air is so dry that the evaporation of perspiration 
cools the body and the heat is not felt to the same extent. 

During the rainy season winds blow in towards the central 
plateau from all sides. The rain falls mainly in thunderstorms, 
which occur almost every day. At the beginning and end of the 
rainy season violent thundery squalls (chubascos) are experienced. 
Rainfall is heavy on the coasts, moderate on the interior plateau 
and very light in the north-west. The skies are cloudy and the 
nights warm and humid. This is the least healthy season. 

The Atlantic coast of Central America has a hot rainy and 
humid climate, the annual rainfall exceeding 120 inches with 
rain in all months. The rain comes mainly in steady falls of 
long duration and wide extent, especially in the winter. In the 
valleys of the interior the rainfall is about 60 inches and comes 
almost entirely in summer, in the form of heavy thunder 
showers. Finally the rainfall increases again towards the Pacific 
coast and averages about 80 inches, though it still falls almost 
entirely in summer and mainly in thunderstorms. 

west indies and Bermuda (Appendix I — Nassau (Bahamas) ; 
Barbados; Havana (Cuba); Port-au-Prince (Haiti); Kingston 
(Jamaica) ; Fort-de-France (Martinique) ; San Juan (Porto 
Rico); Port of Spain (Trinidad); Bermuda). 

The West Indies have a definitely tropical climate, the mean 
annual temperature being everywhere between 75 and 8o° F. 
The Bahamas and Cuba come to a considerable extent under 
the continental influence of North America ; the large islands of 
Cuba and Hispaniola are also large enough to have moderately 
continental climates of their own. Port-au-Prince, on the lee- 
ward side of Hispaniola, has even recorded a maximum tem- 
perature of 101 F., while San Domingo, in the north-west of 
the same island has never exceeded 95 ° F. Southward and 
eastward the climate becomes more oceanic and the small 
islands of the Leeward and Windward groups (Lesser Antilles) 
are entirely dominated by the Trade Winds, and are very 
equable. At Barbadoes, for example, the difference between 
the warmest months (July to September) and the coolest month 
(February) is less than 4 F. compared with n° F. at Havana 
(Cuba) and Nassau (Bahamas). Trinidad and Curacao are 
climatically parts of South America. In winter the Bahamas 
and Cuba occasionally have minima below 55 F. when cold 
waves from the U.S.A. succeed in crossing the ocean. The 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 47 

lowest recorded temperatures are (° F.) : Havana, 50; Nassau, 
51; Kingston, Jamaica, 57; Montserrat, 59; St. Lucia, 60; 
Grenada, 60. 

In the Lesser Antilles the trade winds blow strongly and very 
steadily from east throughout the year. Farther west they are 
weaker, but are steady over the open sea. On the larger islands 
sea breezes develop on the lee sides, especially in the dry 
season, and on the weather sides the trade wind drops to calm 
at night or is even replaced by a light breeze off the land. 
Except in southern groups off the north coast of South America 
the winds also fall light when a storm is passing to the north. 

The rainfall is everywhere considerable; owing to its im- 
portance for the sugar industry rain-gauges are very numerous. 
The annual totals vary greatly according to the exposure; on 
windward slopes they are two, three or more times the fall on 
leeward slopes. In Jamaica, for example, Kingston in the lee 
of the high Blue Mountains has a rainfall of only 31-5 inches 
compared with 137 inches at Port Antonio on the windward 
side. There is a double rainfall season, with maxima in May 
and October, retarded in the southernmost islands south of 
about 1 5 N. to June and November. The driest months are 
February and March, but on windward coasts rain falls through- 
out the year. On the leeward sides of large islands there are 
thunderstorms with heavy rain in April and May. Heavy rains 
also fall when unusually strong winds blow on steep slopes facing 
the sea. On Silver Hill, Jamaica, no less than 135 inches fell 
during 4th-nth November 1909, 30-5 inches in one day. 
Hurricanes are accompanied by torrential rain; in Porto Rico 
11 inches fell in a few hours in August 1899. The floods which 
followed the hurricane of ioth-i3th September 1898 in St. 
Vincent swept away whole villages. The same happened in 
Jamaica in November 19 12. 

Hurricanes can occur in almost any month, but are very rare 
from December to May. They are most frequent between 
August and October. From 1887 to 1923 239 were recorded, an 
average of between six and seven a year, distributed as follows : — 

Jan.-April 



[ay 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


I 


16 


17 


39 


78 


7i 


15 


2 



They usually originate to the east of the Lesser Antilles and travel 
west-south-west at first, usually turning north-east somewhere 



I48 CLIMATE IN EVERYDAY LIFE 

in the West Indian area. The belt of greatest frequency runs 
across Haiti, Cuba and the Bahamas; Trinidad is almost 
outside the hurricane area. The wind speed often exceeds 100 
m.p.h.; in fact, a speed of 130 m.p.h. has been recorded on 
several occasions, and in the storm which destroyed Santo 
Domingo on 3rd September 1930 the speed was estimated from 
the damage as 160 m.p.h. This storm caused many thousands 
of deaths. 

Hurricanes do great damage to crops and buildings, but 
thanks to the efficient warning system of the U.S. Weather 
Bureau the loss of life is generally small. The damage is due 
not only to the winds ; "hurricane waves " flood low-lying ground 
and wash away houses. 

Bermuda has an oceanic sub-tropical climate, but, lying well 
to the north of the West Indies, is cooler and less tropical. Owing 
to the absence of the invigorating trades it is also more ener- 
vating, especially when the south wind is blowing. The winter 
is moderately cool, a minimum temperature of 40 F. having 
been recorded, and even snow is experienced very occasionally, 
lily to October are very sultry months, with hot, oppressive 
nights like a steam bath. The climate is much less healthy than 
that of the Bahamas. Hurricanes sometimes pass over the 
islands. 

The smaller West Indian islands are largely built of porous 
rock, such as coral limestone, and there is no surface water, 
the supply being drawn mainly from wells. In Yucatan, which 
has a similar structure and is similarly situated, the water is 
pumped up by windmills driven by the steady trade winds, and 
there is probably scope for a considerable extension of this 
practice in the West Indies. 

SOUTHERN SOUTH AMERICA 

South America extends from north of the Equator into high 
southern latitudes. Brazil and the northern countries are best 
considered under the head of tropical climates (Chapter VI), 
but most of Chile and the Argentine are sub-tropical or tem- 
perate. As in North America the continent is divided longi- 
tudinally by a high mountain range, the Andes, into two parts 
with very different climates. 

the argentine, Paraguay, Uruguay (Appendix I — Argentine, 



SUB-TROPICAL CLIMATES WITH SUMMER RAINFALL 1 49 

Buenos Aires, Cordoba; Paraguay, Asuncion; Uruguay, Monte- 
video). 

Owing to the shelter of the Andes the eastern part of southern 
South America has a rather continental climate, with con- 
siderable annual and diurnal ranges of temperature, and in 
general a moderate or scanty rainfall. It may be divided into 
a sub-tropical section from 2i°-40° S., and the inhospitable 
Patagonian region south of 40 S., described in Chapter IV. 

The wide plains (Pampas) north of 40 ° S. and east of 65 W. 
have a rainfall of 20-40 inches, increasing to over 80 inches 
locally in the north-east corner. Near the coast the rainfall is 
fairly well distributed through the year, with a maximum 
generally in summer and autumn and a minimum in winter, 
but there is a second minimum at midsummer. Farther west 
the seasonal contrast increases, and west of 62 ° W. the winters 
are almost dry. The weather throughout the year is made up 
of an alternation of cool dry southerly and south-westerly winds 
and warm moist northerly winds. The latter, coming from low 
latitudes, are sultry and enervating. The southerly winds 
generally set in suddenly, bringing the temperature down by 
about 12 F. In spring and early summer they are often stormy, 
when they are known as "Pamperos"; these occur three or 
four times a month from October to January. They sweep 
over the plains, heralded by dust-clouds and bringing local 
heavy showers, which are welcome in summer. On the coast 
dangerous south-easterly gales blow up two or three times a 
year. "Buenos Aires" is rather a misnomer for the city; Don 
Pedro de Mendoza, who named it, happened to arrive on one 
of the rare calm days. It is hot in summer and humid in winter, 
but on the whole it is healthy and very fertile. 

Inland the rainfall decreases and agriculture becomes pre- 
carious. The dry summers are interrupted at intervals by 
sudden downpours which flood the ground; changes of tem- 
perature are large and rapid, and night frosts are frequent. At 
San Luis, in 33 S., the recorded extremes of temperature are 
102 and 1 9 F. Hail is also frequent and plagues of locusts are 
an additional trouble. Along the eastern margin of the Andes 
is a narrow belt of semi-arid country, with few settlements, 
which widens southwards into Patagonia. 

northern and central chile (Appendix I — north to south , 
Arica, Antofagasta, Valparaiso, Santiago). 



I5O CLIMATE IN EVERYDAY LIFE 

The long narrow coastal strip west of the Andes has an extra- 
ordinary diversity of climate. Northern Chile is completely 
rainless ; the average annual fall is less than an inch as far south 
as 29!°, and a year or two may pass without a drop. The rain- 
less regions are watered by melting snow from the mountains, 
and the settlements strung along the rivers are famous for their 
dried fruit. The coast is almost uninhabited, and the road and 
railway run along the inland side of the desert. Between 29J 
and 32 S. a few short, heavy showers fall each year. In 32 S. 
the average is 5 inches, and southward the amount increases 
rapidly to 10 inches in 33 , 20 in 34°, 30 in 35 , 40 in 37 , 
60 in 38 and 80 in 39 S. 

In the interior of Chile the rainfall is generally greater than 
on the coast, but the amount is everywhere less than 10 inches 
as far as 30 S. From about 32 to 42 ° three zones can be dis- 
tinguished, a coastal zone; the valleys between the coastal 
mountains, which are drier than the coast; and the main ridge 
of the Andes, with a relatively high rainfall and a good deal of 
thunder. Between about 35 and 38 S. there are long periods 
of heavy rain ("temporales") which flood the river banks and 
do much damage to the crops along the valleys. In central 
Chile the rain is limited to the winter months; in 30 S. summer 
is almost rainless, but southwards the summer rainfall increases 
and in 50 S. the amounts are almost uniform throughout the 
year. 

The prevailing winds are from south-west in the north and 
from west in the south, very stormy south of 40 ° S. (the "roaring 
forties"). North of 40 ° there are strong land and sea breezes 
blowing up the mountains by day and down them by night, 
and these keep the summer temperatures moderate. 

The climate of Santiago in 33J S. is favourable, with calm, 
mild sunny winters and cool dry summers, so that the gardens 
are very rich. There is an " Indian Summer" of fine warm hazy 
weather at the end of March or beginning of April. The 
northern part of Chile is subject to earthquakes. These are 
generally submarine, and cause great sea waves along the whole 
coast. The mountain climate of the high Andes is discussed in 
Chapter VII. 



CHAPTER VI 
TROPICAL CLIMATES 

MOST of the equatorial regions have a hot, humid and 
rainy climate throughout the year, with very little 
difference between the warmest and coolest months. 
Thunderstorms are frequent and severe, but the equatorial 
regions are almost free from major climatic catastrophes such 
as hurricanes and tornadoes. Moreover, the heat, though 
steady, is never extreme ; over much of the equatorial zone the 
temperature never reaches ioo° F. and rarely falls below 6o° F. 
at moderate altitudes. The heat and humidity are very ener- 
vating, and the absence of a cool season allows no period of 
recuperation; hence whites need a periodical return to a cooler 
climate. Without careful hygiene tropical diseases are rife, and 
vegetation grows so rapidly that it is difficult to cope with. 
Metals corrode, cloth, paper and leather go mouldy. The 
equatorial regions are unsuitable for factory work, but are 
valuable sources of raw material. The only equatorial regions 
which are suitable for permanent occupation by whites are the 
highlands. 

There is still some controversy about the effect of tropical 
climates on the health of white settlers and officials, some 
writers believing that with proper care the tropics are as healthy 
as any other part of the world, and others that they are quite 
unsuitable for white colonisation. The truth probably lies 
between the two extremes. The death-rate where known is 
found to be little if any higher than in temperate countries, 
but this is readily accounted for by the facts that whites going 
to live in the tropics are initially sound and healthy, that they 
live under more comfortable conditions than the bulk of the 
population in the mother country, and that many of the older 
settlers and officials return home on retirement. Even under 
these favourable conditions the sickness rate is high. Sir Aldo 
Castellani (1938) quotes figures for Kenya, which is among the 
healthiest of the equatorial countries because of its elevation, 
showing that in 1938, out of 1,717 resident officials, 1,462 had 
some period of sickness during the year, though the average 

151 



152 CLIMATE IN EVERYDAY LIFE 

duration was only a week. In the tropics one tends to sur- 
render easily to the first symptoms of disease. The main cause 
of both sickness and death is malaria; there is also a remarkably 
high incidence of appendicitis, which may be due to unusual 
kinds of food. A minor trouble, which also appears during heat 
waves in temperate regions, is slight swelling of arms and legs 
(heat oedema). Anyone going to the tropics should prepare 
for this by taking larger shoes, but apart from some discomfort 
there are no ill effects. 

On the coast in tropical countries land and sea breezes are 
well developed during dry periods. The sea breeze coming in 
during the hottest part of the day brings cooler air and welcome 
relief. The land breeze which sets in suddenly in the evening 
or at night is also cool ; whites who have been long on the coast 
and whose vitality has been lowered even find it cold and dread 
it because it causes chills. The body loses some of its power of 
rapid adjustment to changes of temperature and is especially 
liable to chilling by wind. 

The equatorial climate is found in excelsis in the great rain 
forests of the Congo and the Amazon valley. In East Africa, 
Ceylon and the East Indies the climate is intermediate between 
the rain forest and monsoon types. 

THE CONGO BASIN AND NEIGHBOURING PARTS OF WEST AFRICA 

(Appendix I — Belgian Congo, Eala, Elizabethville, Leopolds- 
ville, Angola, Loanda, Mossamedes; French Equatorial Africa, 
Libreville). 

The Congo Basin has a uniformly hot, humid climate. The 
highest temperatures come about the time when the sun is 
overhead at noon, but are never extreme. The cloud cover and 
the forest make the region a perpetual hot-house, very ener- 
vating to whites. The country may be divided into five regions : 

(1) North of about 2 N. there is a short "dry" season from 
December to February, but although there is little rain the air 
remains humid and clammy, and a thick, wet mist (cacimbo or 
"smokes") forms evening and morning. In the far interior, 
where the forests give place to grassland, the natives burn the 
grass during the dry season, and the smoke makes the air very 
hazy and dirty. Immediately after sunset during the dry 
season, over the whole basin, a strong squally wind springs up 
from west, often carrying dust, and reaches a velocity of 20-25 
m.p.h. In the rainy season, which extends from April to 



TROPICAL CLIMATES 1 53 

October with a short break in July, morning and evening are 
generally bright and clear. The dry season is most favourable 
for newcomers, but after some years settlers prefer the rainy 
season. 

(2) Near the Equator there is rain throughout the year, mostly 
in the afternoon, and no differentiation into seasons. The 
interior upland is not unhealthy, but the swamps of the lower 
river are fever-ridden, especially near Boma. 

(3) South of about 3 S. the climate is similar to that of the 
northern region, but with the dry season from June to August 
or September. 

(4) In the extreme south (Katanga, io° S., 26 E.) is a drier 
and healthier plateau climate. 

(5) Along the coast regular land and sea breezes bring 
welcome relief from the heat. Over the whole region thunder- 
storms are very frequent and "tornadoes" similar to those of 
West Africa also occur. 

east Africa (Appendix I — Kenya, Mombasa, Nairobi; 
Nyasaland, Zomba; Tanganyika, Daressalam; Uganda, Entebbe; 
Zanzibar) . 

Owing to the rugged topography and great range of elevation 
the climates of East Africa are very varied. The north coastal 
plain north of about 7 S., about 100 miles wide, has a hot 
climate which is relieved by the large diurnal range of tem- 
perature and humidity. The rainfall is not excessive (40-50 
inches a year on the mainland, about 70 inches on the islands 
of Zanzibar and Pemba) . The worst times are from December 
to March, when the air is uniformly hot, and the period of 
heavy rain in April and May. The pleasantest time is from the 
end of June to mid-September. 

Tanganyika and Nyasaland have a much longer but less 
intense rainy season from November or December to April ; it is 
only near the northern end of Lake Nyasa that the rainfall is 
heavy. Central Tanganyika east of Tabor a is semi-desert. 

Kenya has two rainy seasons, about November and from 
April to June; the latter are known as the "long rains" and 
make travel difficult. From December to March the days are 
hot and sunny, but with some heavy afternoon showers. July 
to September are cool, sunny and dry, sometimes rainless; on 
the highlands the nights are cold and dense mists occur in the 
mornings. 



154 CLIMATE IN EVERYDAY LIFE 

The lake region of Uganda has plentiful rainfall, especially 
on the northern shores of Lake Victoria. The rainfall is well 
distributed through the year, but afternoon rains are especially 
heavy in April to May and August to October. Vegetation is 
very rich near the lake; farther from the shores the rainfall 
decreases. 

Northern Kenya and north-eastern Uganda are hot and arid, 
with a rainfall of less than 20 inches a year which falls mainly 
in short, violent thunderstorms. These regions are of little 
economic value and almost uninhabited. 

Those parts of East Africa which have sufficient rainfall are 
very rich agricultural regions, and the plateau is cool enough 
for permanent settling by whites. Where in a normal year the 
rainfall is barely sufficient, severe droughts sometimes occur, 
and there is a risk here that clearing the bush may alter the 
character of the rains, making them more violent and spasmodic 
and so of less value, and allowing the rain-water to run to waste 
more readily. 

the northern half of south America (Appendix I — Bolivia, 
Sucre; Brazil, Manaos, Para, Pernambuco, Rio de Janeiro; 
Guiana, Georgetown; Columbia, Bogota; Ecuador, Guyaquil; 
Peru, Lima; Venezuela, Caracas). 

The climate of Central America and northern South America 
east of the Andes is mostly typically equatorial, with high but 
not excessive temperatures, a small annual range and high 
humidity. The climate of most regions is enervating and the 
equatorial parts tend to be unhealthy, especially in the lower 
Amazon valley; the drier and more elevated regions are 
healthier. Health is largely a matter of hygiene and drainage ; 
The Panama Canal Zone, for example, was formerly deadly, 
but is now healthy. Cayenne is said to be very unhealthy, but 
this cannot be due entirely to the climate, as Georgetown 
(British Guiana) has a similar climate, but a much better 
reputation. The rainfall is heavy over most of the area, but 
there are "islands" of drought, such as those of Ceara and 
Maracaibo, where agriculture is precarious. Southern Brazil 
extends into the region of sub-tropical climate, with cool in- 
vigorating winters. On the north and east coasts the trade 
winds blow freshly throughout the year, but in the interior the 
winds are light. Land and sea breezes are well developed on 
the coast of southern Brazil ; at Rio de Janeiro they blow very 



TROPICAL CLIMATES 1 55 

regularly, the Bay being protected from the strong north and 
south winds. The sea breeze comes in between noon and 2 p.m. 
and lowers the temperature by 7-1 o° F., bringing relief from 
the heat. In summer thunderstorms frequently occur in the 
evening. The best time of day is the morning, when the sky is 
clear and the air fresh; the worst time in the interior is the 
afternoon, especially during the rainy season, before the after- 
noon rain, when the air is very oppressive. 

The annual variation of rainfall differs according to locality, 
but as a rule the rainfall is heaviest when the winds are lightest 
and most irregular in direction. In Guiana and Venezuela the 
rainy season extends generally from May to November. In the 
north there is a single maximum in June and July; the dry 
season is very pronounced. On the llanos south of the high 
ground the months of December to February are rainless and 
in many years the first rain does not fall until May. The average 
relative humidity is below 60 per cent, and the sky is very clear. 
In the valley of the Orinoco, however, the humidity is high and 
the rainfall considerable in all months. In the west of Vene- 
zuela there is a double maximum; Merida receives i-6 inches 
in February, 10-9 in May, 4-6 in July and 10-4 in October. A 
similar distribution occurs in the upper Amazon valley. 

Near the coast of Guiana the maximum falls in May, when 
the rainfall is very heavy; about September there is little rain. 
The south-eastern coast of Brazil south of 20 S. has a maximum 
in December to March and a minimum from June to August, 
July being almost rainless in places. The coastal plain near 
Pernambuco, on the other hand, has a maximum in June and 
July and a minimum in October to December. The rain comes 
in steady falls of long duration, rather than in the usual tropical 
showers, and thunder is rare, but the amount is very variable 
from year to year. 

In southern Brazil, between latitudes 15 and 22 ° on the 
coast and farther south in the interior there is a definite maxi- 
mum in December to March and an almost rainless winter, but 
south of 22 on the coast, including Rio de Janeiro, there is 
little annual variation. 

Thunder is frequent over most of tropical South America 
east of the Andes, exceeding 100 days a year in the south of 
Matto Grosso and part of the Amazon estuary. The only parts 
in which thunder occurs on less than thirty days a year are 



I56 CLIMATE IN EVERYDAY LIFE 

British Guiana, the middle Amazon valley and the north- 
eastern corner of Brazil, where the coast from Cape San Roque 
almost to Bahia is almost free of thunder. 

West of the Andes the rainfall is very heavy (exceeding 200 
inches) at the foot of the mountains near the Equator, but 
decreases rapidly southwards ; the coast of Peru is almost rain- 
less (less than 2 inches). The temperature on the coast is low 
for the latitude, increasing inland even at moderate elevations. 

EAST INDIES, MALAYA AND PACIFIC ISLANDS (Appendix I — 

Borneo, Sandakan; Celebes, Menado; Java, Batavia; Malaya, 
Singapore; Philippines, Manila, Surigao; Sumatra, Medan, 
Padang; Timor, Keopang; Caroline Islands, Yap; Fiji Islands, 
Suva; Hawaii, Honolulu; New Caledonia, Noumea; Samoa, 
Apia; Solomon Islands, Tulagi; Tahiti, Papeete). 

The climate of the island regions of south-eastern Asia and 
the Pacific is characterised by uniform heat, high humidity and 
abundant rainfall, except in the lee of mountain ranges where 
dry winds occur, such as the bohorok of eastern Sumatra during 
the west monsoon. North of the Equator the winds are north- 
easterly from November to March and south-westerly from 
May to October. In the East Indies south of the Equator they 
are north-westerly (the "west monsoon") from November to 
March and south-easterly (the "east monsoon") from May to 
October. In the transition periods the winds are generally 
light and, on the coasts of the larger islands, are dominated by 
land and sea breezes. On the small Pacific islands west of about 
150 E. the trade winds blow throughout the year. 

There is not much annual variation of temperature ; in the 
north the hottest period is between June and August and the 
coolest January and February. In the south the hottest months 
are January to March and the coolest July and August. Near 
the Equator there are two maxima, about May and October. 
Excessive temperatures are found only in the interior of the 
larger islands, especially in the Philippines, where a range 
from 1 12°-54° F. has been recorded at Tugueguaro in Luzon. 

The rainfall is heavy and falls mostly in thunderstorms, 
which are very numerous and severe, the average frequency 
reaching 322 days a year at Buitenzorg in Java. Near the 
Equator rain falls almost uniformly through the year, but 
farther north and south, in the mountainous islands, there are 
marked wet and dry seasons. The rains fall when the wind is 



TROPICAL CLIMATES 1 57 

onshore ; in the north the eastern sides of the islands are rainy 
from October to February and the western sides from May to 
October. In the south this distribution is reversed. 

The climate of small islands well exposed to the winds is 
pleasant and healthy, improving with distance from the 
Equator. Cocos-Keeling Island is said to be the healthiest 
place in the tropics. Sheltered places in the lowlands of large 
islands are enervating and rather unhealthy, but there are 
good mountain health resorts, such as Tosari in Java and the 
Cameron Highlands in Malaya. 

North of 5 N. typhoons occur, especially in the Philippines, 
which experience four or five severe typhoons each year, mostly 
from July to November (maximum October). When these 
strike Manila or other large towns they do a great deal of 
damage. The extreme south-eastern islands of the East Indies 
are occasionally visited by cyclones between December and 
April. Cyclones are unknown near the Equator, but in the 
Malacca Straits there are violent south-west squalls with 
thunder and torrential rain; these occur only at night, mostly 
between April and October. The idyllic existence of the small 
Pacific Islands is only interrupted at long intervals by the 
passage of typhoons or cyclones. When they do strike an island, 
however, they are disastrous. Notable examples were the Fiji 
cyclones of 3rd March 1886 and 21st January 1904, in both of 
which the smaller islands were overwhelmed by cyclone waves, 
with great destruction not only of houses, but also of coconut 
trees, the islands' most important economic asset. On 15th 
March 1889 Apia, in Samoa, was struck, with the loss of a 
number of warships (but the cyclone may have saved a war!). 
Another cyclone visited Samoa, as well as the Union and Cook 
Islands, in December 1925. Cyclones or typhoons very rarely 
occur between 5 N. and S. latitude, but one wrecked Butaritari 
in 3 N. in the Gilbert and Ellice Islands in December 1927. 

ceylon (Appendix I — Colombo). 

Ceylon has three climates, the west coast, the east coast, and 
the mountainous interior. The south-west and west coasts, 
represented by Colombo, are uniformly hot and moist. The 
annual rainfall is nearly 100 inches, and the mean temperature 
at Colombo ranges only from 84° F. in October to 88° F. in 
March and April ; vegetation is rich. There are two main rainy 
seasons, from April or May to July, following the "burst" of 



I58 CLIMATE IN EVERYDAY LIFE 

the south-west monsoon, and in October and November in the 
transition period between the south-west and north-west mon- 
soons. The intervening period is one of fresh south-west winds 
with clearer skies but occasional bursts of heavy rain. The dry 
season comes in January and February, during the height of 
the north-east monsoon, but in spite of the shelter of the hills 
these months are not rainless. 

On the east coast the rain is less heavy (about 60 inches a 
year), and falls mainly during the north-east monsoon between 
October and January; the remaining months are rather dry 
though not rainless, and the vegetation is much less luxuriant. 
The period of the north-east monsoon is comparatively cool, 
the mean temperature at Trincomalee on the east coast being 
78 F. in December and January, and rising to 85 F. in May 
to July. 

In the mountains the main rainy season extends from June 
to October, during which time the peaks are enveloped in cloud. 
Temperatures at a height of 6,000 feet or so are moderate 
(about 6o° F.), and the hill stations are used as health resorts. 
In Colombo the favourite time for going to the hills is at the 
beginning of the north-east monsoon, when the "land wind" 
sets in, as this is regarded as the unhealthy season on the west 
coast. 



CHAPTER VII 
DESERT, MOUNTAIN AND POLAR CLIMATES 

THIS chapter considers briefly those parts of the earth 
which by reason of drought, elevation or cold are inimical 
to man. They are of direct interest only for mineral 
products, furs and in a few cases as health or holiday resorts, 
but they may be of indirect importance as barriers to trade. 
Their extent is shown in Fig. i . 

desert climates (Appendix I — Asia, Urumtsi; South America, 
Antofagasta, Arica, Lima; Africa, Insalah, Cairo). 

Desert conditions may arise on a small scale as a result of 
soil which is too poor for agriculture, either naturally or from 
overcropping and exhaustion, or because of a very porous or 
fissured sub-soil which allows the rain-water to sink in too 
rapidly. The great deserts of the world, however, are all due 
to a shortage of rain combined with considerable evaporation. 
Outside the polar regions any area with an average rainfall of 
less than 10 inches a year will be unproductive desert unless it 
can be irrigated. 

The significance of deserts in the world's economy is shown 
by the following table of the areas of the desert parts of the 
various continents: — 

Continent Area of desert, square miles Per cent, of continent 

Africa 3,630,000 32 

Asia 1,165,000 7 

North America 458,000 5 

South America 582,000 9 

Australia 1,090,000 37 

Nearly half of this great desert area is made up of the Sahara 
and Arabia, which extend from the Atlantic to the Indian 
Ocean between 15 and 30 N. 

Besides lack of rainfall, the characteristics of deserts are the 
intense insolation, the enormous range of temperature both 
from night to day and from winter to summer, the low relative 
humidity by day and the large evaporation, and the storms of 
dust and sand. In the central Sahara, represented by Insalah, 
and in central Asia (Urumtsi), the averages of the highest and 

159 



i6o 



CLIMATE IN EVERYDAY LIFE 



lowest air temperatures each day, and the daily range, are as 
follows : — 





Insalah 


Urumtsi 




Jan. 


July 


Jan. 


July 


Mean Daily Max. ° F 

Mean Daily Min. ° F 

Mean Daily Range, ° F. . 


69 
40 

29 


117 
82 

35 


3i 
— 21 

52 


94 
50 

44 



The highest and lowest temperatures on record at Insalah are 
I33°F. and25°F. 

The highest temperatures of all are recorded in shallow de- 
pressions where the rocks reflect the sun's heat from all sides. 
In such a situation a shade temperature of 136 F. has been 
recorded at Azizia in Tripolitania — the world's highest — and 
1 34 F. at Death Valley, California. Surfaces exposed to the 
sun reach much higher temperatures by day; at night they cool 
very quickly directly the sun has set. The surface of loose sand 
gets especially hot because of the insulating effect of the air 
between the grains; values exceeding I70°F. have been 
measured (see Chapter VII). Such surfaces are painful to walk 
on, and it has been remarked that in battles soldiers prefer to 
stand up and risk the bullets rather than lie down and be burnt 
alive. Solid rock does not get quite so hot, though a temperature 
of 160 F. has been recorded, but the heat penetrates farther 
into the rock and it cools more slowly. The interiors of tents 
get very hot, and summer dwellings should be very thick walled 
and roofed, or built underground. 

The relative humidity falls to very low figures during the 
hottest days — values of 2 or 3 per cent, are not unusual. The 
drying power of the air depends on the saturation deficit, or the 
difference between the amount of water vapour actually in the 
air and the amount which it could hold if saturated at the same 
temperature (drying power also, of course, increases with in- 
creasing wind velocity). The average saturation deficit in the 
shade at midday at Insalah in July is about 64 grams per 
cubic metre of air, compared with only 9 grams in London. 
A table of saturation deficit in terms of temperature and 
relative humidity is given in Appendix II. At temperatures 
higher than those of the air in shade, such as are reached 



DESERT, MOUNTAIN AND POLAR CLIMATES l6l 

near the surface of the sand, the saturation deficit is even 
greater. 

In spite of the low relative humidity at midday, the daily 
range of temperature is so great that the air is often saturated 
at night and dew is formed. Moreover, the daily change in the 
volume of air is so great that flexible packing covers which are 
not hermetically sealed are soon rendered useless, and the 
contents are subjected to alternate drying by day and damping 
by night. J. Gottmann (1942) remarks that this large daily 
range can be utilised to abstract water from the air by con- 
structing a pyramid of stones. A film of water spreads over the 
stones; as the temperature falls at night the film contracts 
and the surplus water collects in drops which drain away. A 
pyramid of broken limestone with a base 30 feet square gave a 
little water in winter, and in summer more than four pints a 
day. A similar cone constructed in the Crimea is said to have 
given about 80 gallons even on rainless days. The method is 
very old, having apparently been in use in prehistoric times 
(W. Midowicz, 1948). 

The other main characteristic of deserts is the prevailing 
dustiness of the air. Except after the rare rainstorms the air is 
full of a fine haze, which penetrates all crevices. On hot after- 
noons there are whirling pillars of fine sand ; when the wind is 
stronger the whole ground seems to be in motion and walking is 
difficult. But the worst condition is the simoom or "poison wind " 
of North Africa and Arabia, also called chihili or ghibli, which 
is similar to but more intense than the khamsin of Egypt and the 
haboobs of the Soudan. The simoom is a blast of hot air often 
accompanied by heavy clouds of dust or sand, which limits 
visibility to a few yards. The heat is intense (125 F. or more, 
I33°F. has been recorded) and the air seems to glow. The 
dust and dryness not only inflame the eyes, but also cause 
nervous troubles, known in Tripoli as ghiblitis (Castellani, 1938). 
The hot dry winds result in the removal of a large quantity of 
moisture from the body, and this has to be replaced by drinking 
an equivalent quantity of water. As perspiration removes salts, 
these must be replaced, either in the food or by slightly salting 
the water, or heat cramp will result. Since the evaporation of 
perspiration is nature's way of keeping the body temperature 
steady, men who do not perspire freely should keep away from 
all hot dry climates and especially deserts. 



1 62 CLIMATE IN EVERYDAY LIFE 

When the air temperature is considerably above body tem- 
perature and there is appreciable wind, the air brings more 
heat to the body than can be disposed of by the sweat glands, 
in spite of the very low relative humidity. The body tempera- 
ture rises, and if the rise goes far enough, heat stroke ensues 
and in the worst cases leads to death (see p. 22). 

The dust and sand in the air are very injurious to machinery. 
Very little wind is sufficient to raise fine surface sand; R. A. 
Bagnold (1937) states that a wind of 2-5 metres per second 
(5-6 m.p.h.) raises fine dune sand, the quantity transported 
being proportional to the cube of the wind speed above this 
minimum. A. Brun (1944) remarks that the passage of loco- 
motives on railways in desert countries creates a wind which 
fills the air with particles up to a diameter of 1 00 microns (equal 
to one-tenth of a millimetre) . The filter system for ventilating 
locomotives should exclude all particles larger than 50 microns, 
and that of the air admitted to the cylinders all larger than 1 o 
microns, to keep the lubricating oil clean. Ventilation of pas- 
senger coaches also presents difficulty; the best plan is to exclude 
the outside air altogether, and ventilate by air-conditioning 
and fans. The sand frets away paintwork, and a sandstorm can 
strip an automobile clean and polish the bare metal. In Iran 
the summer "wind of 120 days" has even, in the course of 
centuries, undermined walls and buildings. 

The desert imposes a specialised mode of life on its inhabi- 
tants. The motifs of life are heat and water. The heat demands 
loose light clothing, such as the Arab cloak or burnoose and 
baggy trousers; head coverings are essential, hence the turban. 
Except in the larger oases wood is unobtainable, and in the 
settlements houses are built mostly of sun-dried mud bricks or 
adobe. The nomadic Arabs live in tents and follow the scanty 
pasture with their herds; there seems to be a sort of "desert 
telegraph" which spreads the news of where rain has fallen 
and there will be pasture. The Arabs were the first exponents 
of "dry-cleaning," sand replacing water for the ceremonial 
washing enjoined by the Koran. 

In the other desert regions conditions are not so severe as in 
the Sahara, though some of them have even less rain. In the 
deserts of Peru there are places where probably no rain has 
fallen for centuries. In the northern part of the Gobi desert 
the winter is extremely cold, with piercing north winds. In 



DESERT, MOUNTAIN AND POLAR CLIMATES 



163 



less extreme conditions winter in the desert, where water is 
available as in Egypt, is a pleasant, healthy season. 

Where the desert can be irrigated, as in Egypt, Iraq and 
parts of western U.S.A., it is very fruitful, the abundant sun- 
shine and high temperatures giving rich crops, perhaps two 
or three times a year. 

MOUNTAIN CLIMATES 

The characteristics of mountain climates are the decrease of 
air density and temperature and the increase of precipitation 
with height, the distinction between the windward and leeward 
sides of mountain ranges, and the local variety of climates in 
mountain valleys. 

Rough averages of the variations of barometric pressure and 
air density with height are shown in the following table: — 



Table 20. — Decrease of pressure and density with height. 



Height, feet .... 


5,000 


10,000 


15,000 


20,000 


Temperate regions — 










Pressure, mb. 


845 


676 


573 


470 


Density, gm./m 3 


1,070 


880 


770 


660 


Sub-tropics — 










Pressure, mb. 


847 


686 


586 


485 


Density, gm./m 3 


1,020 


850 


745 


640 



Air density is directly proportional to the barometric pressure 
and inversely proportional to the temperature expressed in 
absolute degrees (° C.-f 273). On an extensive plateau on very 
hot days density may be 5-10 per cent, smaller than the figures 
given above. 

Air density affects the performance both of human beings 
and of machines. The effect on humans is inappreciable below 
about 7,000 feet. Above this height a fuller development is 
noticeable in the lungs of the inhabitants, with more oxygen in 
the blood. Above 10,000 feet lack of sufficient oxygen begins 
to cause anaemia and muscular weakness in the permanent in- 
habitants, but even at 14,000 feet large communities exist in 
the Andes without suffering any apparent inconvenience. In- 
experienced mountain climbers suffer from "mountain sickness" 



164 



CLIMATE IN EVERYDAY LIFE 



and sleeplessness, but generally become acclimatised in a 
few days. The effect on machines is most noticeable in the 
decreased power of aircraft to take-off from high-level airfields 
such as those of East Africa, the lift at the same speed being 
directly proportional to the density. This makes a considerable 
difference to the possible load. Air density is also likely to 
affect the acceleration of vehicles driven by internal-combustion 
engines. On the other hand, wind pressure at the same air 
speed is proportional to density, so that at high speed wind 
resistance falls off with increasing height. The decreased density 
of the air on mountains to some extent counterbalances the 
greater strength of the wind often encountered. 

Temperature decreases upwards at the rate of about 3 F. 
per thousand feet, but this does not apply in mountain valleys 
where the lower slopes are often as warm as or warmer than 
the valley floors. The following figures show the mean annual 
temperatures at different heights in southern Peru: — 



Height, feet 
Temperature, 



F. 



Mollendo 

(coast) 



80 
67-2 



La Joya 



4,140 
63-7 



Arequipa 



8,041 
57-8 



Puno 



12,539 
47'5 



Vinocaya 



14,360 
36-7 



El Misti 



19,200 
19-2 



The annual range of temperature is very small, and the climate 
of places like Bogota and Quito, near the Equator at heights of 
eight or nine thousand feet, is often described as "eternal 
spring." But anything eternal can become boring! 

The leeward slopes of mountains are often warmer than the 
windward slopes because of the warm dry winds which blow 
down from the ridge. The Form of the Alps (see p. 108) and 
the Chinook of the Rockies (see p. 125) are the best known 
examples, but similar winds are found wherever conditions are 
suitable. The effect of slope and aspect in the local climate of 
the Alps was described on p. 107. 

The precipitation (rain and snow combined) depends very 
much on the exposure of the mountains to rain-bearing winds. 
On the windward side, especially where the range fronts the 
sea, the precipitation is very heavy, often 200 inches or more a 
year. On the leeward or inland side it is much less. The pre- 
cipitation increases with height up to a level of 7,000-8,000 
feet, but which varies with latitude, being greater nearer the 



DESERT, MOUNTAIN AND POLAR CLIMATES 1 65 

Equator and smaller nearer the poles. With increasing height 
the proportion of the precipitation which falls as snow increases 
and the zone of greatest snowfall is above that of greatest pre- 
cipitation. The most favourable situation for a regular water- 
supply for irrigation or hydro-electric works occurs where there 
is abundant winter snowfall which melts gradually during the 
spring and summer. Heavy falls of rain are dangerous because 
they cause sudden floods in the narrow mountain valleys, where 
the streams may rise 60 feet in a few hours. 

The strength of the sunlight, due to the thin air and the 
reflection from snow surfaces, combined with the bracing 
quality of the cold air and brisk winds and the beauty of the 
surroundings, make accessible mountain districts very suitable 
for sanatoria and for holiday resorts, but otherwise mountain 
districts are of little importance apart from their mineral wealth 
and, on the lower slopes, forest products. 

High plateaus have more extreme climates than mountain 
peaks. The climate of the plateau region of North America 
was described on p. 126. Tibet (Appendix I, Lhasa) at an 
average height of over 12,000 feet, has intensely cold winters but 
hot summers. Eastern Tibet has a considerable summer rain- 
fall during the south-west monsoon and a rich vegetation; the 
rainfall is not heavy, but is persistent ; winters are dry with little 
snow, and the winds are strong and piercing. Western Tibet 
has less precipitation but more snow in winter. 

Arctic climates (Appendix I — U.S.S.R., Dikson Island, 
Verkhoyansk; Alaska, Nome; Canada, Chesterfield, Dawson, 
Hebron; Spitsbergen, Green Harbour). 

In the polar regions the sun remains continuously above the 
horizon in summer and below it in winter, and this as much as 
the cold is the dominant feature in the life of high latitudes. 
Even at midsummer the sun is low in the sky, but the days are 
so long that the amount of heat received is comparable with 
that in temperate latitudes. In continental regions the winters 
are intensely cold, the average temperature of February being 
about — 25 F. with extremes as low as — 6o° F., but the 
weather is generally fine and there is little snow. On flat 
ground the winter is long, but southerly slopes warm up quickly 
in spring, the snow melts early, and the soil temperature is 
much higher than the air temperature; such sites have a local 
favourable climate. The summers are short and cool with an 



1 66 CLIMATE IN EVERYDAY LIFE 

average temperature of about 40 ° F. in July, and gener ally- 
dull ; there are frequent but not heavy showers of rain or snow. 
The great curse of summer, especially in the tundra, is the 
plague of mosquitoes, which are active throughout the twenty- 
four hours. 

On the ice-bound shores of the Arctic Ocean the climate is 
much more severe. The winters, though not so cold as in the 
interior, are long and dreary; the summers are very short (only 
a month or two), cold, foggy and rainy. At the mouths of the 
great rivers flowing into the Arctic Ocean the break-up of the 
ice in spring causes great floods and ice-jams, making the 
country impassable. The most favourable Arctic climates occur 
where the coasts are washed by a warm current, as in southern 
Iceland and south-western Spitsbergen. The east and west 
coasts of Greenland also are not especially cold in winter, and 
have experienced temperatures exceeding 70 F. In West 
Greenland the temperature is extraordinarily variable, and 
may change by 70 F. or more in two or three days. 

Economically the most important part of the Arctic comprises 
Alaska, northern Canada and Labrador (Appendix I — Nome, 
Chesterfield, Hebron). The Arctic zone forms a narrow belt 
along the west and north coasts of Alaska, which broadens 
eastwards, the southern boundary running roughly from 70 N. 
in i6o°W. (north-west Alaska) to 55 N. on the east coast of 
Labrador, and then down the coast to Belle Isle. The climate is 
severe; it is very cold in winter, all waterways and the soil being 
frozen and the tundra covered with snow packed by the wind 
into solid drifts. The coasts are ice-bound for much of the year 
and only accessible for a few months in late summer. Summer 
is short and cool, the temperature only rising above 50 F. in 
the rare warm spells. In the northern and north-eastern part 
of the area the ground is permanently frozen to depths of 100- 
200 feet, only thawing at the surface for a few inches in summer, 
when the surface becomes soft and marshy. The rainfall is 
small, but evaporation is also slight, and the numerous lakes 
and rivers maintain a high humidity. When the thaw comes in 
late spring the country is a quagmire. Western Alaska along 
Bering Strait is somewhat warmer than the remaining part of 
the area, but the climate is damp and foggy, with much snow in 
winter. The neighbourhood of Hudson Bay is also damp and 
foggy in summer, with low day temperatures. The coast of 



DESERT, MOUNTAIN AND POLAR CLIMATES 1 67 

Labrador is especially dismal, owing to the Arctic ice-drift, 
and here the climate improves inland. Over the whole area 
agriculture is only possible in a few specially favoured spots; 
the chief products are minerals and furs. 

Life in the Arctic regions presents many problems, some of 
which have been overcome by war-time research. About 70 
per cent, of the loss of heat from the body occurs by the ordinary 
processes of radiation, conduction and convection, and this loss 
is proportional to the area of the body multiplied by the differ- 
ence between the temperature of the surface of the clothing and 
that of the environment (air and surrounding bodies). Hence 
the clothing must be such as to transmit little of the body heat. 
The right clothing is not so much a question of weight and 
thickness as of adequate insulation by air spaces. Wind adds 
greatly to the effect of the cold, and the loss of heat brought 
about by even a moderate wind is known as "wind chill." In 
the coldest weather it is necessary to keep out the external air, 
which can be done by an impervious outer garment with draw- 
strings at the openings of neck, arms and legs, but it is also 
necessary to permit free circulation of the air next the skin, to 
evaporate perspiration. This can be secured by wearing a 
"string vest" — a coarse wide-meshed net to hold the under- 
clothing away from the body. The main problem is then to 
keep the extremities warm. In a cold environment the body 
automatically tries to reduce its loss of heat by reducing the 
flow of warm blood to the limbs, and the consequent return of 
cooled blood. This helps to maintain the body temperature, 
but at the expense of the fingers, toes, nose and ears. If this 
process goes too far frost-bite results. For this reason it is 
necessary to wear properly designed gloves and footwear as 
well as suitable body clothing. 

The loss of heat from the body has to be made good by eating 
a correspondingly large amount of food, especially fats and 
carbohydrates. For this reason and also because of the diffi- 
culty of raising crops, fish and flesh are the basis of diet. Alcohol, 
which lowers resistance to cold, is to be avoided. A survey of 
the physiological reactions to cold, with a full bibliography, is 
given by B. Roberts (1943). 

Building on permanently frozen ground ("permafrost") is 
difficult, because the heat from the building thaws the ground 
and causes subsidence. The method found most satisfactory is 



1 68 CLIMATE IN EVERYDAY LIFE 

to drive wood or concrete (not metal) piles into the ground so 
that they project one or two feet below the surface to which 
thawing penetrates. The upper parts of the piles should be 
smooth and greased, so that they are not disturbed by move- 
ments of the thaw layer. The floor of the building must, of 
course, be clear of the latter. Water supply is a difficulty, 
especially where the lakes and rivers freeze solid ; where possible 
it is desirable to instal a large tank which can be kept per- 
manently heated throughout the winter. In many Arctic 
regions there is plenty of potential water-power for hydro- 
electric generators, but this is not available in winter when it is 
most required. Transport to isolated communities in Canada has 
been mostly by air, but the Canadian Government has de- 
veloped a " snow-mobile " which has proved satisfactory and is 
being put into manufacture. 



PART II 

CLIMATE AS AN ENEMY 



CHAPTER VIII 

CLIMATE AND THE DETERIORATION OF 
MATERIALS 

IN this chapter we discuss the effects of climate on the physical, 
chemical and organic structure of materials, especially manu- 
factured goods. Mechanical effects are due mainly to ex- 
tremes of temperature and especially to excessive heat. The 
effect of frost on roads and buildings was discussed in Chapter 
II (p. 80). Chemical effects (corrosion) and organic effects 
(growth of mould) are due mainly to the combination of high 
temperature and high humidity. 

SOLAR HEAT AND HIGH TEMPERATURES 

The "temperature" shown on climatic charts is the "shade 
temperature" or temperature inside a "screen," which is a 
box freely ventilated on all sides. The temperature of a body 
exposed to radiation from the sun is generally much higher. 
It depends on the solar radiation, the nature of the surface of 
the body, particularly its colour, heat capacity and conduc- 
tivity; the nature of the surroundings, which determine the 
amount of heat received by reflection and radiation from sur- 
rounding objects; and the loss of heat by conduction to the air, 
which depends on the air movement. Of these factors one of 
the most important is the incoming solar radiation. As ex- 
plained in Chapter II this depends on the elevation of the sun, 
i.e. the latitude and season, and on the purity and dryness of 
the air. Colour is equally important; in hot dry climates a non- 
conducting black surface, such as a cloth, may be 30-40 F. 
hotter than a similar white surface ; coloured surfaces are inter- 
mediate in temperature. In general a matt surface absorbs 
more heat than a polished one, but this effect is less important 
than that of colour. 

The surface of material which is a poor conductor of heat, such 
as wood or^plastic, will become hotter than a metal surface 
which is not insulated from the ground, but it will also be a 
better protector for the space which it covers. The worst 

171 



172 



CLIMATE IN EVERYDAY LIFE 



construction for a store is a single iron roof, of a dark colour, 
supported on non-conducting walls, with no insulating layer 
or ventilating holes below it. _> 

As an illustration of the physical problem involved let us 
consider a thin plywood container, painted black and mounted 
clear of the ground, on a calm, clear day with a vertical sun. 
We will assume (C. E. P. Brooks, 1946^) that the container is 
large enough for the temperature of the ground beneath to 
equal the shade temperature, say a maximum of 50 C. (122 F.), 



Sun 

1 



I SCO 



l/cm 2 /, 



mm. 



Confciinar 




Ground 

Fig. 14. — Radiation to and from a container with thin walls. 

that the radiation on the upper surface is 1 -8 cal./cm. 2 /min., all 
of which is absorbed, and that there is no transfer of heat other- 
wise than by radiation (see Fig. 14). 

Let T G be the temperature of the ground, equal to 50 C. or 
323 A. (temperature in degrees absolute is 273 plus tem- 
perature in ° C). Let T B be the temperature of the lower 
surface of the container and T s that of its upper surface, the 
temperatures of the internal and external surfaces being 
identical in each case (this is approximately true for very thin 
plywood). The radiation from a surface is proportional to the 
fourth power of its absolute temperature. The upper surface T s 
is receiving radiation from the sun (i-8 cal./cm. 2 /min.) and 



CLIMATE AND THE DETERIORATION OF MATERIALS 1 73 

from the bottom surface at the rate of crT^ 4 where a is the 
radiation constant, and is sending out radiation on both sides 
at the rate of aT s 4 . Similarly the bottom surface is receiving 
radiation from the top surface at the rate of aT s * and from the 
ground at the rate of aT G * 9 and is sending out radiation from 
both surfaces at the rate of crTg 4 . Then assuming that each 
surface has reached equilibrium with its surroundings, i.e. is 
radiating as much heat as it is receiving, 

ar/=^r s *+T-/)=^(r s * +3 23 4 ) 
*T s *=±(i-a+*T B *y 

If the ground and plywood surfaces radiate as black bodies, 
(7=82 X io~ 12 gm.cal./cm. 2 /min. The solution of these two 
equations gives 

r s -368° i4.=95° C or 203 F. 

7*=330 A=75 C.ori6 7 F. 

Experiment has shown that the mean temperature of the air 
inside the container will be some degrees lower than the mean 
of the upper and lower surfaces, i.e. between 8o° and 85 C. 
(i8 5 °-i94°F.). 

If the sun is not vertically overhead we can allow for the 
increased absorption due to the longer path of the rays through 
the air by the method described in Chapter II (p. 59), but as 
we are dealing with the hottest, clearest days, and as we have 
to include also the effect of some of the scattered solar radiation, 
a reasonably close value for the transmission coefficient would 
be 0-9. The maximum temperatures of the top and bottom 
surfaces of an empty plywood container in different latitudes, 
calculated in this way purely from the exchange of radiation 
are: — 



Latitude 


0-30 


40° 


50° 


6o° 


Top surface, ° F. 


203 


194 


181 


163 


Bottom surface, ° F. 


167 


158 


H5 


127 



Temperatures calculated in this way are, however, too high. No 
body absorbs all the solar radiation falling on it, and there is 
always some loss by conduction to the air. 

The following table shows some observed temperatures of 



174 CLIMATE IN EVERYDAY LIFE 

non-conducting surfaces exposed to the sun, compared with the 
maximum temperatures in the shade : — 





Latitude 


Temperature ° F. 




Observed 


Shade max. 


Bare sand, Loango .... 
Black cloth, Khartoum .... 
Black soil, Poona, May .... 
Bare sand, Sahara .... 

Aircraft wing, Tucson, Arizona 

Balloon, U.S.A 


5 

(25) 
32 


183 
184 
148 
173 
{215 
U93 
152 


(120) 
118 

(99) 
(120) 

118 

78 



The maximum possible temperature calculated from the 
balance of radiation is about 200 F. for each of the first five. 
The outstanding observation of 2i5°F. at Tucson, Arizona 
(Madison, 1944), is more than 20 F. above the next highest 
figure of 1 93 F. at the same place, and must have been due to 
a very unusual combination of circumstances. Apart from this 
the readings are mostly 6o°-7o° F. above the shade maxima and 
io°-20° F. below the temperature calculated from the exchange 
of radiation with no allowance for loss by conduction. 

In the observations on the surface of a balloon (Washington, 
Bureau of Standards), the incoming solar radiation at the time 
was measured. The surface temperature calculated from it by 
the method described above was 158 F., which is only 6° F. 
above the temperature actually observed. In the Sahara there 
is a good deal of dust haze which reduces the intensity of the 
radiation. At Poona in May there is a good deal of water 
vapour in the air, and also the soil probably had a higher con- 
ductivity than loose bare sand. The readings at these two 
places are therefore relatively low. 

In the upper layers of air inside a closed container heated 
from above the temperature decreases downwards, but the 
lower layers have a nearly uniform temperature equal to or 
very little above that of the bottom skin. The actual distribu- 
tion depends on the size, shape, material and contents of the 
container as well as on the temperatures of the top and bottom 
skins, but the type of distribution is indicated in Fig. 15. If the 
sun is not directly overhead, so that one side of the container 
is in sunshine and the other in shadow, the effect of the unequal 
heating is to raise the surfaces of equal temperature somewhat 



CLIMATE AND THE DETERIORATION OF MATERIALS 



175 



140"F. 



125 °f. 



iTFT 



II2VF 



on the warmed side and depress them somewhat on the cool 
side; if the heating on the two sides is very unequal a circulation 
of air will probably be set up which will tend to equalise the 
temperatures throughout the container. The average tem- 
perature inside the container is lower than the mean of 
the top and bottom surfaces, especially if the sun is nearly 
overhead. 

The temperatures reached by solid conducting bodies such as 
steel rails are probably similar to or 
slightly lower than the average inside 
temperatures of containers exposed 
in the same way, but comparable 
data are few. A steel rail painted 
black, exposed at Panama, reached 
a temperature of 129 F. with a 
shade air maximum of 88° F. (H. G. 
Gornthwaite, 1920). 

The "black-bulb" thermometer in 
vacuo is often used as a measure of 
the highest temperature which would 
be reached by an object exposed to 
the sun. This instrument consists 
of a self-registering maximum ther- 
mometer, the bulb of which is 
covered with lamp-black and en- 
closed in a larger glass bulb which 
is exhausted of air. The readings 
of these instruments are rather un- 
reliable, depending on the size of the 
thermometer bulb, the constitution 
of the black coating, the composition 
of the glass sheath and the com- 
pleteness of the vacuum, but in general they record tempera- 
tures about equal to those found inside large unventilated 
containers, and much lower than those of non-conducting 
upper surfaces. 

Fig. 16 is an attempt to estimate the distribution of the 
highest temperatures which would be reached in an average 
year by horizontal non-conducting surfaces exposed to the sun. 
It is based on such actual observations as are available, on 
readings of black-bulb thermometers increased by 20 F., and 



hot 



Fig. 15. — Temperatures 

inside a container heated 

on top surface. 



176 CLIMATE IN EVERYDAY LIFE 

on the mean annual maxima of shade air temperatures; the 
latter were charted by G. E. P. Brooks and G. L. Thorman 
(1928). The readings are considered to be substantially inde- 
pendent of height above sea-level, sin£e the greater intensity 
of solar radiation with increasing height is balanced by the 
lower air temperature and consequent greater loss by con- 
duction. 

The mechanical effects of high temperature may be increased 
by the chemical effects of strong sunlight and especially of ultra- 
violet radiation. These are rather obscure, but according to 
W. M. H. Schulze (1941) strong UV radiation causes lacquer 
to become brittle, coloured fabrics to bleach and soft rubber to 
crack. UV radiation is especially strong in clear, dry air at 
high levels. 

Effect of colour. — We owe to G. W. Grabham (1921) a series 
of valuable experiments on the effect of colour. The general 
result of these was that black and dark objects took . up. higher 
temperatures than white or light-coloured objects. Grabham 
expressed his results by taking the difference between the tem- 
peratures of black and white objects as 100 and expressing the 
temperature of any object of another colour as a percentage of 
this difference. For our purpose, however, it seems better to 
express the difference between a black object and the cor- 
responding shade maximum temperature as 100, and the excess 
of a coloured object over the shade maximum as a percentage 
of this. Table 2 1 shows the effect of colour, calculated in this 
way, based on the results of a number of experiments in hot 
countries. Grabham's experiments include: the temperature 
under one thickness of thin cloth, with a number of thicknesses 
beneath; and thermometers inserted through corks in small 
cylindrical tin flasks laid on a doubled woollen blanket. Experi- 
ments with thermometers in square flat tins were carried out 
by W. F. Harvey (1930) at Kasauli in the Punjab (quoted by 
Sir A. Castellani, 1938). The experiments with steel rails were 
made by H. G. Cornthwaite (1920) and those with different 
coloured soils at Poona by L. A. Ramdas and R. K. Dravid 
(1936). The latter gave the weekly mean temperatures of 
the surfaces at 2 p.m.; corresponding shade temperatures are 
not available and the figures at the foot of the table are the 
mean daily shade maxima in January and May at Poona. 
Grabham's figures for cloths and painted flasks incorporate 



CLIMATE AND THE DETERIORATION OF MATERIALS 



177 




I78 CLIMATE IN EVERYDAY LIFE 

experiments on several different days, but the results are in 
general very consistent. 



Table 2 1 . — Temperatures of different colours as percentages of 
"black" minus "shade max." 





Cloth 


Painted tins 


Steel rail 


Soil- 


Poona 




Colour 














Mean 




Khartoum 


Haifa 


Kasauli 


Panama 


Jan. 


May 




Black, ° C. 


84-3 


70-8 


58-0 


53'9 


(53) 


64-4 




F. 


184-0 


i59-o 


136-0 


129-0 


(127) 


148-0 






0/ 

/o 


/o 


/o 


/o 


% 


/o 


% 


Black 


100 


100 


100 


100 


100 


100 


100 


Dark blue . 


89 












8q 


Brown 




88 


— 


— 


— 


— 


88 


Cement wash 


— 


8^ 


— 


— 


— 


— 


85 


Green 


— 


79 


— 


9° 


— 


— 


85 


Plain metal 


— 


73 


9° 




— 


— 


81 


Grey, ash colour 


— 


79 




— 


70 


76 


75 


Khaki, light 
















brown, yellow 
















brown 


76 


72 


— 


— 


69 


78 


74 


Red . 


— 


6q 


— 


66 


77 




73 


Pale blue . 


72 












72 


Straw 


— 


54 


— 


— 


— 


— 


54 


Cream 


— 


49 


— 


— 


— 


— 


49 


White 


42 


40 


54 


58 


42 


40 

(30) 


46 


Shade max, ° C. 


42 


42 


34 


31 


(30) 


(37) 




°F. 


108 


108 


93 


88 


(86) 


(99) 





These figures may be summarised roughly as follows to show 
the relative effect of colour : — 



Black ......... 100 

Dark blue, brown, green ...... 85-90 

Grey, cement wash, ash, plain metal . . . 75-85 

Khaki, red, light brown, pale blue, aluminium paint . 70-75 

Pale colours (straw, cream) ..... 50-55 

V White ......... 40-50 

Boxes with glass sides and tops gave about the same tempera- 
tures as blackened tins. Different whites react very differently; 
e.g. in one experiment with painted flasks the standard white 
gave a percentage of 39 and white enamel one of 49. The 
white-painted steel rail in Panama appears to have given an 
abnormally high temperature for this colour; on the other hand, 
in another series of observations at Poona in January 1934 a 



CLIMATE AND THE DETERIORATION OF MATERIALS 1 79 

surface of powdered white chalk gave a percentage of only 
about 30 (added in brackets in the table) . The range shown by 
other colours is also no doubt due mainly to differences in the 
shade of colour and the texture of the surface. Unpainted, 
horizontal steel rails reached the following temperatures at 
Agra, India, and in Hungary: — 





Max. temp, of rail 


Shade max. 


Agra, India 
Hungary 


6i°C. (i42°F.) 
53 G. (127 F.) 


44° C. (m°F.) 
33 C. (91 F.) 



Harvey remarks that the high temperatures reached by water 
in closed receptacles are lethal to micro-organisms, while those 
of water open to the air are incubation temperatures. He also 
found that sprinkling water at 45 C. (ii3°F.) on a tin at 
45 G. every ten minutes reduced the temperature by as much 
as io° C. (i8°F.). 

The most important effects of high temperature acting alone 
appear to be chiefly mechanical, such as expansion, buckling of 
metal rails, blistering of paint work, but the effects on viscosity 
or rigidity, such as weakening of glued joints, softening of 
gelatine films, etc., are scarcely less important. In some 
materials, such as rubber, there is a critical temperature above 
which a non-reversible chemical change takes place. The most 
important effect on rubber however is oxidation under the 
influence of strong sunlight (J. Crabtree, 1948). 

The speed of most chemical reactions increases with tem- 
perature at an increasing rate. "Dunn's equation" is in the 
form 

logA=-C/T+K 

where A is the speed of reaction and T is the absolute tem- 
perature; C and K are constants. For the range of action met 
with in nature it is convenient to express the rate at o° C. 
(32 F.) as unity. Then writing t=T— 273 (i.e. t in ° C), we 
have 

A-a'l T . 

The chemical reactions with which we are concerned are 
negligible below — io° C. and are mostly concentrated in the 
higher ranges of temperature met with in nature. The effective 
range of t is therefore small compared with 273+f, where t is of 



l8o CLIMATE IN EVERYDAY LIFE 

the order of 30 G. Thus a sufficient approximation is given 
by replacing T by its mean value and then writing 

A=cl 1 where 0L=a 1,T . 

The ratio of cd to <x' +1 ° is termed the temperature coefficient of 
chemical reaction or </> 10 , and is generally found to lie between 
2 and 3. Actually log</> 10 is inversely proportional to T (T-\- 10) 
and decreases as temperature rises. It varies, however, with 
different materials ; for corrosion of iron in pure dry air it is 
only about 1-2 (data from W. H. J. Vernon, 1935). For most 
actions involving water it appears to be very close to 2. 

Flow resulting from high temperature may be of importance. 
From such few figures as I have been able to find it appears 
that the fluidity F of viscous substances can be represented with 
fair accuracy by an expression of the form 

F=axb< 

similar to that for chemical effects. Examples for two sub- 
stances are: — 

a b "Temperature coefficient" 

Glycerine 0-02 1-09 2-4 

Pitch 2Xio -12 1-277 11 '5 

It is well known that many organic actions, such as the 
growth of bacteria, increase with temperature in much the 
same way as chemical actions. There is generally an optimum 
above which the activity falls off very rapidly (sterilising effect) , 
but this optimum is in most cases above the maximum tem- 
perature likely to be experienced in the natural environment. 
It is believed that the temperature coefficient is in general 
about 2, i.e. the rate of bacterial action doubles for each rise of 
temperature by io° C. (18 F.), but it varies considerably with 
the temperature. D. Snow, J. A. B. Smith and N. G. Wright 
(1944) found that on feeding stuffs impregnated with nitrogen 
compounds, moulding appeared after 128 days at 15 G. and 
after 60 days at the optimum temperature of 22 G. (tem- 
perature coefficient 2-95). At 37 G. there was no moulding. 
On the other hand, figures given by D. Snow, M. H. G. Crichton 
and N. C. Wright (1944) for locust beans at 75 and 80 per cent, 
relative humidity at temperatures of 15-5 and 20 G. give a 
value for the temperature coefficient of only 1 -4. 



CLIMATE AND THE DETERIORATION OF MATERIALS l8l 

A full discussion of the effect of high temperatures on electrical 
apparatus is given by W. M. H. Schulze (1941), and this should 
be consulted for details. Heat may affect the insulation of 
electrical apparatus, and all insulating materials for use in hot 
countries should be stable up to at least 90 C. (194 F.). 



T£HE EFFECT OF HUMIDITY 

Many organic materials react to the relative humidity of the 
air, absorbing moisture when the humidity is high and releasing 
it when the humidity is low, with corresponding changes of 
volume. The final result is independent of the air temperature, 
though the rate of reaction increases with temperature. This 
property of expansion with increasing relative humidity is made 
use of in the "hair hygrometer," which measures the relative 
humidity by the lengthening or shortening of a bundle of 
human hairs. ,, pighrejative humidity leads to warping and 
swellin g of wo od, p aper, lea ther, etc., low relative humidity 
to shrinking and cracking. In very humid climates and in 
climates with a large diurnal range of temperature any machine 
components such as electrical connections which are sensitive 
to humidity should if possible be enclosed in airtight water- 
proof covers of glass, porcelain or plastics. 

The effect of humidity on the physical properties of paper has 
been investigated by F. T. Carson (1944). With increasing 
humidity paper expands by different amounts in different 
directions, causing distortion and difficulties of registration in 
colour printing. At low humidities paper becomes brittle and 
readily cracks along folds. 

The reaction of paper, leather and similar materials to 
increasing humidity often increases rapidly when the relative 
humidity rises above 65 or 70 per cent. With some materials, 
such as hair, the absorption is increasingly rapid with increasing 
relative humidity through the whole range from o to 100 per 
cent. In others, such as leather, there are three stages : ( 1 ) from 
o to 15 or 20 per cent, absorption of water is moderately rapid; 
(2) from 20 to about 65 per cent, absorption is slow; (3) above 
65 per cent, absorption increases more and more rapidly. A 
typical curve (average of curves for untanned hide and various 
leathers given by J. R. Kanagy, 1947) is shown in Fig. 17. 

Many organic materials, such as tobacco, sugar and cereal 



1 82 



CLIMATE IN EVERYDAY LIFE 



products, glue, etc., which contain moisture, have a critical 
relative humidity at which they remain in equilibrium and 
retain their properties. At higher humidities they absorb 
moisture, and at lower humidities they dry out, both processes 
often being accompanied by deterioration. It is important to 
ascertain this critical humidity and ensure that it is maintained 
in the store. 

H. W. Eades (1945) found that the rate of rusting of cans 





1 » 


v - 




r * 1 












O 30 


















<0 










L. 










5 















































a 




J,.. 




1 1 



IO 20 30 40 50 6O 70 80 90 IOO 

Rcl&rive humidify 

Fig. 17. — Absorption of water- vapour by leather. 

stored in wooden boxes increased directly with the moisture 
content of the wood above 17 per cent.; below this figure 
rusting was slight and below 14 per cent, practically absent; 
the species of wood was unimportant. Dehumidification, e.g. 
by silica gel, prevented rusting. 

If kept for too long in a high relative hmnidity many organic 
substances go mouldy even at moderate temperatures. In most 
cases the critical humidity is 65-70 per cent. Fig. 18 shows 
a number of curves of rate of growth of mould in air streams 
of different relative humidities, given by D. [Snow and colla- 
borators (1944, 1945) and by N. G. Wright (1944). The papers 



CLIMATE AND THE DETERIORATION OF MATERIALS 1 83 

were kindly sent to me by Dr. Wright. The data are given as 
the number of days before formation of mycelium and before 
fructification; to get the rate of moulding I have taken the 
reciprocals of these and expressed them as percentages of the 
value at 80 per cent, humidity. The curves are: — 

(1) Cereal feeding stuffs with urea mixture (moulding) 
(Snow, Smith and Wright). Time at 80 per cent. R.H. 
21 days. 




70 73 SO 85 SO 95 IOO 

Relative humidify per cent. 
Fig. 18.- — Rate of growth of mould at different humidities. 

(2) Artificially dried grass. Mycelium (Wright) 18 days at 

80 per cent. 

(3) Ditto, fructification (25 days at 80 per cent.). 

(4) Locust beans, 15*5° C, mycelium (Snow, Crichton and 

Wright) (22 days at 80 per cent.). 

(5) Ditto, 20 G. (19 days at 80 per cent.). 

The figures for oats were stated to be similar to those for locust 
beans. 

The curves are in general agreement, and it is seen that the 
growth of mould does not begin until the relative humidity 
exceeds 64 per cent., increases slowly to about 70 or 72 per cent, 
and then rapidly. It is probable that the rapidity of increase 
with increasing humidity falls off above about 90 per cent. 



184 CLIMATE IN EVERYDAY LIFE 

Snow and Wright ( 1 944) found that loss of dry matter from 
stored bran due to enzymatic and microbial activity at 20 G. 
increased slowly as humidity rose from 64 to 84 per cent, and 
very rapidly above 84 per cent. 

THE EFFECTS OF COMBINED TEMPERATURE AND HUMIDITY 

Many chemical and organic actions require the presence of a 
moist atmosphere, and apart from any purely thermal effect the 
rate presumably depends on the quantity of water vapour 
available. In saturated air the moisture content nearly doubles 
for each rise'of temperature by io° C; actually the coefficient 
is not constant, but decreases from 2-0 between — io° and o° C. 
to 1 -68 between 30 and 40 C. Between 20 and 40 C. the 
weight of water vapour e in grams per cubic metre can be 
represented on the average by 

*=5*9X 1-054' 

where t is the temperature in °C. The effect of water vapour 
depends, however, not only on the quantity present in the air, 
but also on the willingness of the air to part with it, i.e. on the 
relative humidity. In general, actions involving water vapour 
do not occur with a relative humidity below 60-70 per cent., 
though the figure probably varies widely. The combined 
effects of high temperature and high humidity may be divided 
into chemical and organic effects. 

The main chemical effects of high humidity at a high tem- 
perature are rusting and tarnishing of metals. In experiments 
on the rusting of iron W. H.J. Vernon (1935, 1945) found that 
action was very slow until the relative humidity reached about 
65 per cent., and then increased rapidly at an almost linear rate 
with increasing relative humidity, until it was retarded by the 
formation of a protective crust. N. Cabrera and J. Hamon 
(1947) found that in pure air the oxidation of aluminium at 
constant temperature was appreciable in dry air, but increased 
with relative humidity at an increasing rate. In dry air oxide 
is formed, in damp air hydroxide. In the presence of ozone the 
increase with humidity was very rapid when the relative 
humidity was above 50 per cent. 

Organic effects such as rotting and mildewing also depend 
on botrTltemperature and humidity; mould growths, for example, 



CLIMATE AND THE DETERIORATION OF MATERIALS 1 85 

do not occur with relative humidities below about 70 per cent., 
but with high temperature and saturated air their growth is 
very rapid. 

"The "general expression for both chemical and organic effects 
involving the presence of water vapour at a high temperature 
may be put in the form 

. (H-k) . t 

A=b- (1*054) 

100 v °^ J 

where A is the rate of action, H is the relative humidity, / is the 
temperature in ° C, b and k are constants. An alternative form is 

A=b (V-V t ) 

where V is the observed vapour pressure and V t is the vapour 
pressure at the same temperature t but with relative humidity k. 

It is to be remarked that H is not necessarily the relative 
humidity in the free air, but depends on the amount of water 
vapour in the air, the surface temperature of the body being 
attacked, and any additional sources of water vapour. Damp 
wood in an unventilated environment, for example, may 
saturate the air in contact with it and provide a suitable medium 
for fungoid growth, even where the free air is below the critical 
humidity. 

The rate of action also varies with time in a complex way. In 
some cases, e.g. corrosion of metals, a protective crust is soon 
formed, after which the action slows down. In organic decay, 
on the other hand, the rate of action may speed up with time 
as the surface of decay or concentration of bacteria or fungus 
spores increases. 

Experiment showed that the sum of the values of 
1-054^//— 65)/io for the twelve months has a value of about 
100 in the worst tropical conditions. I have therefore adopted 
this expression as a standard index number to define the 
rate of deterioration of materials due to temperature and 
humidity. Values were calculated for a number of places with 
suitable averages of temperature and humidity, and a chart 
based on these is shown in Fig. 19. In using this chart it must 
be remembered that in islands and places of high relief the 
index may vary rapidly from place to place; in a world chart 
it is possible to present only a broad generalisation. The chart 



1 86 



CLIMATE IN EVERYDAY LIFE 




CLIMATE AND THE DETERIORATION OF MATERIALS 1 87 

does not take account of atmospheric impurities which greatly 
increase chemical deterioration (see Chapter VIII). 

Corrosion of metals can be prevented to some extent by 
coating the surface with protective varnish, but any crack or 
pit in the varnish allows corrosion to spread rapidly beneath it. 
The coating may be broken by channelling due to rain ; a more 
insidious process is the adherence of small animal or vegetable 
fragments to the surface. These go mouldy and the mould in 
time attacks the varnish, setting up centres of corrosion/ 

The values of the deterioration index charted in Fig. 1 7 take 
no account of the diurnal variation of temperature and humidity, 
but this may be important. Although chemical and organic 
effects are retarded at low temperatures a large diurnal range 
of tem perature may cause fresh s upplies of moisture to enter 
a container which is not completely air-tightT The daily range 
in a container exposed to radiation by day and night may 
exceed 40 C. This means that more than 10 per cent, of the 
contained air is expelled during the day and replaced at night 
by fresh air which may be nearly or quite saturated. Influx of 
this moist air may well result in the deposition of free moisture 
within the package. It is not unlikely that the clotting of finely 
ground material is largely due to this daily interchange of air. 
The effect is especially marked if the material is at all hygro- 
scopic. The corrosion of metals and the deterioration of 
materials affected by moisture are accelerated by the deposition 
of dew, which is very heavy in the open in some tropical 
countries, especially where there is a large daily range of tem- 
perature. Dew forms even in dry regions such as North Africa 
and centra l India where the deterioration index given bv the , 
monthly values oftemperature and humidity is small or zero. 
and this probably accounts for the deterioration and corrosion 
which is sometimes observed there. 

Condensation often occurs in transit when the goods are 
loaded in bulk in a cold climate and later exposed to warm 
moist air. Water collects on flat surfaces, and H. W. Eades 
(1945) recommends packing of tins on their sides to avoid 
formation of pools on the flat tops. Lacquering is a partial cure 
for rusting, but the flanges or rims form weak spots ; dehumidi- 
fication of the store or cargo space of ships is the best preventive. 



CHAPTER IX 
ATMOSPHERIC POLLUTION 

MAN is himself responsible for one of his greatest troubles, 
the pollution of the atmosphere by the products of com- 
bustion, especially the burning of coal. These are in- 
jurious to the health of man and animals and to plant life; they 
attack stone and metal, and the deposit of soot makes it diffi- 
cult to maintain standards of purity in products such as con- 
fectionery. It has been estimated (A. Parker, 1945) that about 
one-fortieth of the coal burnt in domestic grates goes up the 
chimneys ; factories are much better, but even there nearly one- 
eightieth of the coal burnt goes into the air. A detailed account 
of the effects of smoke on health, plant life, buildings, etc., and 
of methods of prevention is given by A. Marsh (1947), based 
mainly on the work of the Atmospheric Pollution Committee 
of the Department of Scientific and Industrial Research. 

Atmospheric pollution takes the form of tar, ash and other 
solid particles, some of which may be hygroscopic, and in- 
jurious gases, especially sulphur dioxide. The corrosive effect 
of impurities is greatly magnified by high humidity. Thus 
W. H. J. Vernon (1935) showed that at 99 per cent, relative 
humidity air with o-oi per cent, of S0 2 is thirty-five times as 
corrosive of iron as is pure air. Moreover, the effect of S0 2 is 
enormously magnified by the presence of particles such as 
charcoal, and if these are screened off rusting is slow. Particles 
of charcoal in conjunction with S0 2 are by far the most corro- 
sive; curiously, chemically active particles such as ammonium 
sulphate are less dangerous, but these and even particles of 
silica increase the action of S0 2 to some extent. Apart from 
that the effect of acid impurities increases nearly in proportion to 
their concentration. The concentrations of S0 2 met with in the 
open rarely exceed ixio -5 per cent., or one part in ten million. 
In the open the quantities of water vapour (at a given tem- 
perature and humidity) and impurities available for corrosion 
depend on the wind velocity. The effect of wind in increasing 
corrosion can probably be represented by the factor (i+bW) 
where W is the wind speed in miles per hour, and b may pro- 

188 



ATMOSPHERIC POLLUTION 1 89 

visionally be taken as 0-067. The effective wind velocity is the 
air movement at the surface of deterioration, and in enclosed 
spaces with only a very slow interchange with outside air, W is 
practically zero. 

The combined effect of temperature, humidity, atmospheric 
impurity and wind may be expressed by the general equation 

A=a^ t +b ^ H '~ 65 \ i-o^y(i+cI)(i-{-o-o6yW) 

where A is the rate of deterioration of a fresh sample, t is the 
temperature in ° G. at the surface of deterioration, H is the 
relative humidity at temperature t, I is the concentration of 
effective impurities, bacteria or fungus spores, and W is the 
effective wind velocity in m.p.h. ; <2, a, b and c are constants to 
be determined in each specific case. When H is less than 65 per 
cent., H— 65 is taken as zero for surfaces affected by moisture. 

Another cause of deterioration in the open is rain containing 
impurities, especially SO s . It is possible that in some cases this 
completely outweighs that due to water in the form of vapour. 

Hints on the minimising of corrosion due to atmospheric 
pollution are given by J. G. Hudson (1948). One recommenda- 
tion is that paint should as far as practicable be applied when 
the relative humidity is below 70 per cent., i.e. in suitable dry 
weather for outdoor painting and in heated rooms for indoor 
painting. Flame cleaning, followed by painting while the 
surface is still warm, is also a good method for existing badly 
corroded structures. Acid-resistant enamels are the best pro- 
tection against an atmosphere polluted by sulphur dioxide. 

The action of atmospheric impurities on building materials 
has been described by R.J. Schaffer (1938-9). There are two 
main effects. First, soot, especially tarry matter, adheres to 
the surfaces, causing discoloration of walls and loss of light 
from windows and reflecting surfaces. This involves expense in 
frequent painting and cleaning. Secondly, sulphur dioxide and 
ammonium sulphate attack the carbonates of lime and magnesia 
which make up limestones and magnesian limestone. The Houses 
of Parliament are constructed of magnesian limestone, which 
the atmosphere of London converts into Epsom salts. Siliceous 
materials are not attacked, but calcareous sandstones, which are 
very durable in the country, disintegrate quickly in large towns. 
The soluble sulphates and carbonates formed by the action of 



I90 CLIMATE IN EVERYDAY LIFE 

SO 2 and CO s on limestone are brought to the surface when 
the stone dries off after rain and form crusts. Where these are 
broken decay proceeds rapidly and unsightly holes are formed. 

In addition to its effect on buildings, dirt in the air cuts down 
the light and heat received from the sun, especially in winter, 
and this weakens the health of town-dwellers. Fig. 20 shows 
a cross-section of London, comparing the duration of bright 
sunshine in winter with the deposits of atmospheric pollution. 
This shows that London's coal consumption cuts down the 
sunshine in the centre of the city by about 100 hours in the 
three winter months. Over the year as a whole the loss is 
about 300 hours. In addition there is a considerable loss 
of heat even when the sun appears to be shining strongly. 
At Leicester (A. R. Meetham, 1948) the loss of ultra-violet 
radiation in winter was found to be almost exactly proportional 
to the amount of atmospheric impurity. A striking indictment 
of coal smoke as an enemy of society is set out in the pamphlet 
"Guilty Chimneys," issued by the National Smoke Abatement 
Society. 

THE DISTRIBUTION OF ATMOSPHERIC POLLUTION 

The main sources of atmospheric pollution are the large 
towns, both factory chimneys and the open fires of residential 
buildings contributing to it. From the towns it is carried by the 
winds. The coarser solid particles mostly fall to the ground 
within a mile or so of the source, the finer solid particles are 
carried much farther but eventually sink to the ground under 
their own weight, and the gases remain in the air until they are 
brought down by rain, but at a distance from the source their 
concentration is so weakened by diffusion that they become 
innocuous. For many years measurements of atmospheric 
pollution have been made by means of "pollution gauges" at 
a constantly growing number of places in Great Britain. These 
are large porcelain funnels with a surface area of four square 
feet, which collect the material falling into them or carried in 
by rain; the deposits accumulate in a large bottle and are 
analysed monthly. The measurements are published in the 
Annual Reports of the Advisory Committee on Atmospheric 
Pollution, published by the Department of Scientific and In- 
dustrial Research. They are given in metric tons per square 
kilometre; one metric ton weighs 2204-6 lbs. and is therefore 



ATMOSPHERIC POLLUTION 



191 



— — r~ — - — r~— 

8 



R 



P 



\J0\/A\19^JQ 




?(sys\ 



o 

1 



o 



o 

1 

-1 



§ 

§ 
8 



bo 



■\u>("bs/s>d04 ^i/sodap pips 
1 l 



192 



CLIMATE IN EVERYDAY LIFE 



slightly less than the English ton of 2,240 lbs. One metric ton 
per sq. km. equals 2-56 English tons per square mile or 9 lbs. 
per acre. These gauges have been installed mainly in large 
towns, and so far we have no information about the atmospheric 
pollution in Wales, south-west England, Kent and East Anglia. 
The distribution of solid precipitates in Britain, and those 
brought down by rain, was studied on the basis of these reports 
by C. E. P. Brooks (1948). The total amount of matter involved 
is very large, as is shown by the following table for England : — 



Table 22. — Atmospheric pollution in England. 





Area, 
square miles 


Deposits, 
tons per sq. 
mile per year 


Total deposits, 
tons 


Products of 
coal, tons 


London 

91 other towns 

Country 


425 

1,490 

47,800 


260 

200 

85 


1 1 1 , 1 00 

301,080 

4,088,700 


92,400 

235,460 

1,982,400 


Total 


49.715 


9i 


4,500,800 


2,310,260 



The total deposit over the country as a whole is made up in 
nearly equal proportions of the products of coal combustion 
and of other material such as road dust, organic matter, smoke 
of bonfires, etc. In open country the "other material" which 
I have termed "country pollution," is estimated to account for 
slightly more than half the total deposit, but in towns the great 
bulk of the pollution is due to the incomplete combustion of coal. 
The general distribution of deposits over the country is shown 
in Fig. 2 1 , which is based on the measurements of pollution 
gauges either well away from large towns or to the west or 
south-west of them. It shows three main areas of pollution, in 
the Clyde Valley, the industrial belt of Lancashire and West 
Yorkshire, and the London area, with subsidiary centres in 
Tyneside and near Birmingham. The distribution of pollution 
in towns could not be shown on this map because of the very 
local nature. In the centre of a large town and for about a mile 
to the eastward the annual deposits may amount to from 130 
to over 200 tons per sq. km. or from 340 to over 500 tons per 
square mile. From this dirty centre the amounts decrease 
slowly eastwards and rapidly westwards and south-westwards. 
A composite map showing the distribution around a number of 



ATMOSPHERIC POLLUTION 



'93 




Fig. 21. — Sketch map to show distribution of total solid deposits 
(metric tons per sq. km.) per annum over open country and to 
windward of main centres of pollution. 



*94 



CLIMATE IN EVERYDAY LIFE 



large towns as a percentage of that in the centre is shown in 
Fig. 22. In any individual town the distribution is modified by 
the topography and by the presence of local sources of pollution. 




Fig. 22. — Composite map of pollution expressed as a percentage of 
that in the centre of a town. 

Thus in the Glasgow district (Fig. 23) the area of maximum 
pollution is a long narrow ellipse which follows the valley of 
the Clyde, but there is a subsidiary centre in the south. 

The detailed distribution of pollution in the neighbourhood 
of a town depends on the local sources. It could probably be 



ATMOSPHERIC POLLUTION 



195 



worked out roughly by a study of these, or a rapid survey could 
be made by means of petri dishes and a portable apparatus for 
measuring suspended impurity (Anon., 1 944) . For a description 
of a planned survey of atmospheric pollution in and around 
Leicester, see A. R. Meetham (1948). 

The concentration of smoke particles in the air (as distinct 
from the deposited particles) increases from 7 mg./ioo cubic 




Fig. 23. — Distribution of pollution in Glasgow. 

metres in Leicester (population 260,000) to 84 mg. in London 
(population 8,000,000), but local conditions have much influence. 
Thus Cardiff, where low-volatile coal is burnt in domestic grates 
instead of the usual bituminous coal, is a very clean city. 

The smoke from a large town affects the country for a 
considerable distance to leeward. I estimated that even in the 
large area of Greater London only about one-third of the smoke 
produced is deposited within the urban area; the remaining 
two-thirds is carried into the countryside or even out to sea. 
London smoke is noticeable as far away as Norwich. The 



I96 CLIMATE IN EVERYDAY LIFE 

proportion of the smoke of smaller towns carried into the coun- 
try is naturally much larger, probably about nine-tenths. An 
analysis of the composition of the deposited pollution in different 
localities showed that the ash and carbonaceous particles are 
largely deposited near their sources, and the soluble material 
is mostly carried into the country. 

The amount of pollution held in suspension in the air is also 
of great interest. "Suspended impurity" is measured by draw- 
ing a known quantity of air through filter paper and comparing 
the resulting stain with a standard set of stains ; measurements 
are made automatically every hour at a few places in Britain. 
The results at Kew Observatory, Richmond, were discussed by 
H. L.Wright (1932). 

Town air contains a very large number of microscopic smoke 
particles — as many as 870,000 per cubic inch (53,000 per cubic 
centimetre) have been found during a dense fog in London. 
The average number at Kew Observatory in 1928-30 was 
12,500 per cubic inch (760 per cc), but the annual variation is 
very great, ranging from 39,000 per cubic inch in December to 
only 1,000 in June. The number was three times as great with 
winds from east (i.e. from London) than from other directions. 

A town fog brings out the full unpleasantness of suspended 
atmospheric pollution. A clean white country fog, though it 
reduces visibility and is damp, is not otherwise unpleasant. 
In a large town the same fog is a dirty brown, often with an 
acrid taste and smell, and this is entirely due to the admixture 
of atmospheric impurities. The latter include a number of 
hygroscopic nuclei on which the water vapour condenses. This 
condensation increases the weight of these particles, which tend 
to fall out of the air. G. M. B. Dobson (1948) points out that 
for this reason a town fog becomes much cleaner during the 
night, when the outpouring of smoke into the air is least, but 
becomes dirty again as soon as fires are lit in the morning. 
The fog droplets also dissolve the sulphur dioxide present in 
polluted air, forming sulphurous acid which gradually oxidises 
into sulphuric acid. This tends to prevent the fog droplets 
from evaporating when the relative humidity falls, and Dobson 
considers that this may be the reason why fog often persists 
longer in towns than in the country. 

The diurnal variation of suspended impurity is of interest in 
connection with the ventilation of buildings. H. L. Wright's 



ATMOSPHERIC POLLUTION 197 

figures for Kew Observatory are shown in Fig. 24 (average for 
1928-30). This shows that the air is cleanest about 3-4 a.m. 
and 2-3 p.m., and twice as dirty about 9 a.m. and 8 p.m. The 
diurnal cycle shows an interesting parallel with the cycle of 
human activity. In the early morning domestic fires are out 
and factory furnaces are banked. After 6 a.m. the fires are lit 
or stoked with great production of smoke. Following on this 
morning activity the fires burn more cleanly, but the greatest 
factor in the afternoon minimum is probably the greater tur- 
bulence, resulting in the more rapid mixing of the smoke-laden 




IO 12 a.m. Z 4- 6 8 MO Mid. 

Time, GMT. 
Fig. 24. — Suspended impurity at Kew Observatory. 

air with cleaner air from above. This turbulence decreases 
during the evening and the lighting-up of home fires produces 
the second maximum from 7-10 p.m. 

Atmospheric pollution is a serious trouble in the manu- 
facturing areas of the United States. There is a well-known 
rhyme about Pittsburgh: — 

"Mary had a little lamb, 

Its fleece was white as snow; 
She took it down to Pittsburgh, 
And look at the poor thing now!" 

Note. — A serious disaster occurred recently at Donora near Pittsburgh. Donora 
lies in a narrow winding industrialised valley, from which the smoke finds difficulty 
in escaping during periods of calm anticyclonic weather. Such a situation persisted 
from 26th to 30th October 1948, and was accompanied bv thick persistent fog. 
According to a preliminary report by R. D. Fletcher (1949), near the end of 
this five-day period the accumulation of "smog" (a graphic portmanteau word 
for "smoky fog") was so stifling that hundreds of people were affected and 
twenty died. 



I98 CLIMATE IN EVERYDAY LIFE 

In this city the steep valleys form conduits through which 
atmospheric pollution flows by air drainage from the numerous 
industrial centres in the Ohio Basin (H. F. Hebley, 1948). 

Air which has travelled for a long distance over the sea is 
generally very clean, and on west coasts in regions of prevailing 
westerly winds the towns benefit from this supply of clean air. 
In California where, because of the cold current off-shore, there 
is often an increase of temperature with height up to a "ceiling" 
at a height of a few hundred feet, the smoke produced by the 
towns is not dissipated by turbulence, and pollution is a serious 
problem in the centres and eastern parts of these towns. The 
problem in Los Angeles, for example, was discussed by C. G. P. 
Beer and L. B. Leopold (1947). On east coasts with prevailing 
westerly winds the neighbourhood of the sea makes little differ- 
ence in winter, but in summer sea breezes bring in cleaner air 
by day. The local distribution of smoke near the source and 
the means of minimising the nuisance were described in 
Chapter II (p. 92). 



MINIMISING ATMOSPHERIC POLLUTION 

Efficient methods of combustion can do a great deal to 
decrease the emission of smoke into the air, but they cannot 
eliminate it altogether. They are less successful with the 
emission of gaseous impurities such as sulphur dioxide 
which escape with the flue gases, though in some large plants 
such as the Battersea Power Station about 90 per cent, of 
the sulphur dioxide is removed from the effluent gases by 
washing. 

The problem of control of pollution is a perennial one, but 
the most thorough investigation arose in connection with 
claims for damage to crops in the State of Washington by gases 
from the smelter works at Trail in British Columbia, seven 
miles from the boundary in a deep, narrow valley. The results 
of this investigation are described by E. W. Hewson (1945). It 
was found that in light winds along the valley the gases, which 
are initially about 70 F. warmer than the surrounding air, rise 
to a height of about 500 feet above the chimney stacks, and 
then flow along the valley in a broad ribbon which is initially 
about 200 feet deep. The spreading of this layer of polluted air 
along the sides and bottom of the valley is brought about 



ATMOSPHERIC POLLUTION 1 99 

partly by the local circulation of air in the valley and partly 
by eddy diffusion due to turbulence in the air. 

The local circulation depends on the local topography, and 
was investigated by aircraft and kite-balloon as well as by 
observations on the ground, the sulphur dioxide itself being 
used as an indicator. The turbulence was measured by a 
special instrument devised for the purpose, described in detail 
by Hewson. The effect of turbulence on a layer of gases emitted 
at a considerable height above the ground shows a marked 
diurnal variation. In the early morning the temperature near 
the ground increases upwards (inversion) and the air is stable. 
Consequently the gases remain at about the height of emission, 
where they spread out to form a definite layer. Shortly after 
sunrise the ground begins to warm up and a turbulent layer 
forms near the ground, destroying the inversion. As soon as the 
top of this turbulent layer reaches the level of gases the latter 
spread rapidly downwards to the ground. This is the most 
dangerous period. An hour or two later the inversion is com- 
pletely destroyed, the turbulent layer extends well above the 
level of emission of gases, and the gas layer is dissipated upwards 
and quickly disappears. 

The control measures adopted depend largely on the wind 
direction and speed and the degree of turbulence. When the 
wind is more than 5 m.p.h. and is blowing away from the agri- 
cultural area, emission of gases can do no harm. When the wind 
is less than 5 m.p.h. or is blowing towards the agricultural area, 
emission is only reasonably safe when the turbulence at the 
level of emission is sufficiently great to dilute the gases with 
clean air and so lower the concentration before they have 
travelled far enough to do damage. These control measures 
are checked by actual observations of the concentration of SO a 
at a point some distance down the valley from Trail. 

Hewson points out that similar measures of control could be 
adopted to regulate the emission of gases in level country, with 
the additional advantage that emission could be planned in 
advance in accordance with the weather forecasts, which is not 
practicable at Trail because of the mountainous nature of the 
country. 

When the emission takes place at or near ground level, as in 
the cleaning of sewage plants, the first stage in the dissipation 
is omitted. So long as there is an inversion of temperature the 



200 CLIMATE IN EVERYDAY LIFE 

gases will tend to drift over the ground at a low level, but when 
the lower air warms up the turbulence will quickly carry them 
upwards. Gardeners in residential districts who wish to remain 
popular with their neighbours will also do well to consider 
whether or not there is an inversion before lighting a smoky 
bonfire. 

SALT NUCLEI 

In addition to smoke particles, dust, organic debris, etc., 
there are present in the air ultra-microscopic particles which 
are hygroscopic and act as nuclei of condensation for water 
vapour. Some of these originate as products of combustion, but 
many of them are believed to be molecules of salt derived from 
the evaporation of sea spray. These are especially numerous in 
coastal districts and, during gales with the wind on-shore, salt 
particles may be carried far inland. They probably account 
for some chemical corrosion and have also been known to 
interfere with electrical transmission by forming a conducting 
layer over the surface of insulators. A notable case occurred 
in October 1927, following a south-westerly gale (Anon., 1927). 
This carried a considerable amount of salt spray inland. In 
South Wales the salt solution immediately caused shorting and 
failure of transmission. At the time, however, the air was dry, 
and the spray soon evaporated, leaving salt crystals. These 
were plastered over the insulators of power lines in the Mid- 
lands, but the salt was too dry to be a good conductor, and there 
were no transmission failures there until a day later, when the 
humidity rose nearly to saturation point. 



CHAPTER X 



CLIMATIC ACCIDENTS 



I 



N this chapter we discuss briefly the nature and distribution 
of some of the more violent atmospheric phenomena. These 
may be listed as follows : — 



(i) Exceptionally heavy rains. 

(2) Floods. 

(3) Hail. 

(4) Snow. 

(5) Ice storms. 

(6) Lightning. 

(7) Tornadoes and squalls. 

(8) Hurricanes, typhoons and tropical cyclones. 

Frosts were dealt with in Chapter II and exceptionally high 
temperatures in Chapter VIII. 



EXCEPTIONALLY HEAVY RAINS 

In the design of buildings, roads, airfields, etc., it is essential 
to take account of the greatest rainfall to be expected in a short 
period. The heaviest bursts of rain last only a few minutes; as 
the time interval considered grows longer the average intensity 
falls off. The heaviest rainfalls which have been recorded any- 
where in the world in different periods of time are : — 



Time 
Minutes 


Amount 
inches 


Rate 
in./hr. 


Place 


Time 
Hours 


Amount 
inches 


Rate 

in./hr. 


Place 


1 

5 

14 

60 

60 


1-02 
2-28 
3 9 

IO'O 

n-5 


6i-o 
27-4 
167 

IO'O 

"•5 


California 

Panama 

Roumania 

New South Wales 

California 


3 

4 

24 

2 days 

4 days 


16 
17 
46 
66 
102 


5-3 
4-3 
1-9 
1-4 
II 


Pennsylvania 

Queensland 

Philippines 

Formosa 

India 



These may be termed "freak rains"; the odds against en- 
countering a fall of these amounts in any particular place in 
any one year are so great (probably millions to one) that the 
risk need not be taken into account. We are concerned here 
with falls that may reasonably be allowed for. 



202 



CLIMATE IN EVERYDAY LIFE 




CLIMATIC ACCIDENTS 203 

Fig. 25 shows the maximum rainfall to be expected in one 
hour once in two years. This map is based on a number of 
actual records of autographic gauges and on statistical relations 
of the rainfall in one hour to that in a day, and to the mean 
annual number of thunderstorms. The construction of the map 
was described by C. E. P. Brooks and N. Carruthers (1946), 
who also give a map of the maximum to be expected in two 
hours. Fig. 25 shows that once in two years a fall of 2-5 inches 
or more in an hour may be expected at places in the wetter 
parts of Central America, around New Orleans, the Cameroons 
and coast of Nigeria, Natal, eastern Madagascar, the west 
coast of India, Malaya and Java, around the Gulf of Tongking, 
and north-west Australia; locally the fall may exceed 3 inches. 
As an indication of the intensity of such a fall it may be re- 
marked that it has probably been exceeded only once any- 
where in the British Isles since records began. 

For various purposes we require to know the maximum fall 
to be expected in different intervals of time in different numbers 
of years. The following table gives the highest falls to be 
expected at any one place in different periods in the British 
Isles (based on E. G. Bilham, 1936) and the Gulf Coast of 
U.S.A. (based on D. L. Yarnell, 1935): — 



Table 


23-- 


-Heaviest rain in 


different periods, 


inches. 


Once 


Britain 


Gulf Coast of U.S.A. 


years 


5 


30 


1 


2 


24 


5 


30 


i 


2 


24 




mins. 


mins. 


hour 


hours 


hours 


nuns. 


mins. 


hour 


hours 


hours 


2 


0-24 


o-45 


o-6 


0-7 


i-6 


o-55 


i-8 


2-4 


3-2 


5'4 


5 


0-36 


0-62 


o-8 


I'O 


2-0 


o-6o 


2*2 


2-9 


4-0 


b-4 


10 


o-45 


0-77 


1 -o 


1-2 


2-5 


o-66 


2-4 


3*2 


4'5 


7-8 


25 


0-62 


1-02 


«'3 


i-6 


3*2 


0-77 


2-7 


3'7 


5-3 


9-b 


50 


0-77 


1-25 


i-6 


i'9 


40 


0-87 


3-0 


4-1 


6-o 


io-6 


100 


0-92 


1-50 


2'0 


2-4 


4-8 


I -00 


3'3 


40 


6-6 


14-0 



In Java the highest falls in forty-six years were (inches) : 5 
minutes, 0-63; 30 minutes, 2-3; 1 hour, 3-7; 2 hours, 5-6; 
24 hours, 1 1 -4. 

For Britain E. G. Bilham (1936) gives the expression: — • 

n=i.25*(r+-i)- 3 ' 55 



204 CLIMATE IN EVERYDAY LIFE 

where n is the number of falls in ten years equal to or exceeding 
r inches in t hours. This may be written 

£=(-i25r#)° ,282 -o-i 

where R is the maximum rainfall to be expected in H hours 
once in T years. For other parts of the world, especially the 
U.S.A., other formulae have been devised. One of the simplest, 
given by R. W. Powell (M. M. Bernard, 1932) takes the form 

R=c(TH)* 

i.e. the maximum rainfall to be expected in H hours once in T 
years is proportional to the fourth root of the product of the 
number of years by the number of hours. In Fig. 25 T=2 
and H=i } so that the maximum rainfall to be expected in H 
hours once in T years is 0-7 times the value read off the chart, 
multiplied by the product of the square roots of the number of 
years and the number of hours. Note that the latter may be a 
fraction, i.e. 30 minutes is f hour. 

It is simplest, however, to consider the number of years 
and the time interval separately. For the former Brooks and 
Garruthers (1946) give the simple approximation that the maxi- 
mum to be expected in T years is (i+log T) times the average 
maximum in one year. If T is 2, (i-flog T) is 1-3; if T is 
10, ( 1 -flog T) is 2, and so on. This rule gave good results. 
In Fig. 25 T is 2, so that the factors by which figures read off 
the chart are to be multiplied to give the maximum rainfall 
in an hour expected in T years are: — 



r 


1 


2 


5 


10 


25 


50 


100 


:tor 


077 


i-o 


i*3 


i'55 


1-85 


2-1 


2*3 



Correction to other intervals of time is more difficult, since 
no simple expression seems to cover adequately the whole range 
of time from a few minutes to twenty-four hours. 

For rainfalls in short periods of, say, five minutes to one hour, 
a favourite expression with engineers takes the form 

R=aT/(T+b) 

where R is the rainfall (inches) in T minutes; a and b are 
constants, but the values assigned to them vary widely. In 
Britain and similar temperate climates a safe upper limit is 



CLIMATIC ACCIDENTS 205 

given by taking a as 7/6 of the value for one hour read off 
Fig. 25, and b as 10. In tropical and sub-tropical regions the 
heaviest falls are more persistent, so that the ratio of the maxi- 
mum fall in five minutes to that in one hour is smaller. Upper 
limits to maxima in these regions are given by taking a as 4/3 
of the value read off Fig. 25 and b as 20. 

For periods of one to twelve hours Brooks and Garruthers 
(1946) found that a good approximation is given by the rule 
that the maximum to be expected in N hours is (i-flog N) 
times the maximum to be expected in one hour. The form 
R=aNI(N J r i) adopted by B. D. Richards (1944) gives identical 
values for durations of one to four hours, but lower values for 
durations of six and twelve hours. The ratios given by these 
two expressions are shown in the following table : — 



Time, hours . 


1 


2 


4 


6 


12 


Ratio 
i+logJV 
aJV/(JV+i) . 


i-o 
i-o 


i-3 
i-33 


i-6 
i-6 


i-8 
17 


2-1 
I-8 5 



The maximum in a day, expected once in two years, is shown 
in Fig. 26. For conversion to the maximum in some number of 
years other than two, the factors given on p. 204 apply. It 
is to be remarked that Fig. 26 refers to the maximum rainfall 
in the "rainfall day," generally from 9 a.m. to 9 a.m. next 
day, and not to the maximum in any period of twenty-four 
hours. As, however, the maxima generally come in thunder- 
storms, which are most severe in the afternoon and evening, the 
difference is not great. 

The most intense falls are generally confined to relatively 
small areas. In severe thunderstorms in the British Isles the 
average duration at any place is about two hours, and the 
average area over which rain is falling at the same moment at 
the rate of an inch or more in an hour is about 20 square miles 
(50 square miles in exceptional storms), while the area over 
which rain is falling at the rate of 2 inches per hour ranges from 
less than one to three square miles. As the storms move over 
the country the actual area of heavy rain during the day is 
much greater than this. Thundery rain of less intensity generally 
falls over much wider areas. In the tropics the areas of very 
heavy rains are probably larger, but there are few data. 



206 



CLIMATE IN EVERYDAY LIFE 



-8 



m 



°S 'b °8 ^ "8 




o 

o 

o 
<u 

Oh 

X 



O 



a 



CLIMATIC ACCIDENTS 



207 



J. Glasspoole 
severe storms: — 



1930) quotes the following areas for typically 



Table 24. — Areas of rainstorms. 





Rainfall exceeding (inches) 




6 


5 


4 


3 


2 


I 


London, 1 6th June 191 7 
Angerton, Northumberland, 7th 
September 1898 


sq. mis. 

3 


sq. mis. 
II 


sq. mis. 

0-4 
27 


sq. mis. 

4*4 
50 


sq. mis. 
20 


sq. mis. 
51 



A number of figures for different parts of the world, for 
storms lasting a few days, are given by B. D. Richards (1944), 
for areas up to 6,000 square miles. His figures for eastern 
U.S.A., West Australia and Central India for storms lasting 
on the average four days show that, if the square mile of maxi- 
mum rainfall is taken as 100 then the averages over larger 
areas centred on this are : — 



Area in sq. miles 
Rainfall intensity 



1 

[00 



500 1,000 2,000 4,000 6,000 
89 85 79 71 67 



In Britain, according to figures given me by J. Glasspoole, the 
intensity falls off more rapidly with increasing area, being 72 for 
an area of 500 square miles and 51 for 3,000 square miles. 



FLOODS 

The risk of flooding by heavy rain depends on the nature of 
the ground as well as the amount of rain. A valley with a small 
catchment area is more liable to sudden floods than a river 
system with a large basin, because the areas of most intense 
rain are local, and usually affect only a part of a large basin; 
moreover, in the latter rain falling in different localities reaches 
the main river at different times. For these reasons the maximum 
discharge to be expected from river basins of different areas 
(but otherwise similar) is proportional, not to the area, but to 
the area raised to some power less than one. The maximum 
flow Q,may be put in the form Qj=cA p R where A is the area 
and R the maximum rainfall. Estimates of p vary from 2/3 to 



208 CLIMATE IN EVERYDAY LIFE 

5/6; 3/4 may be taken as an average figure. The constant c 
depends on the steepness, permeability, etc., of the basin. 

The risk of flooding cannot be calculated directly from the 
probable maximum rainfall, because the loss by percolation 
into the ground, ponding, evaporation and absorption by 
vegetation is very variable. The actual run-off from a single 
heavy fall may vary from about 10 per cent, to nearly 100 per 
cent, of the fall; it probably averages somewhat less than 50 
per cent. Fortunately in summer, when the heaviest thunder- 
storms occur, the ground is generally most able to absorb a 
large proportion of the water. The floods most destructive to 
property occur in the larger river basins, such as the Mississippi, 
but these rise slowly and in the Mississippi floods, thanks to the 
flood-warning system of the U.S. Weather Bureau, there is 
rarely much loss of life. The modern flood forecasting service 
was described by M. Bernard (1948). Smaller rivers, especially 
in hilly country, rise much more rapidly and are more liable 
to cause loss of life. The Ohio is especially dangerous. 
According to J. B. Kincer (1937), in the years 1903- 
35 the damage to property by floods in the U.S.A. amounted to 
about 550 million pounds (excluding soil erosion) and the loss 
of life to about 3,000. For details of some great floods in the 
United States see pp. 121, 123. 

In any particular situation the heights of floods in the past 
are the best guide, provided that the bed of the river has not changed. 
Some rivers are unstable, frequently changing the shape of their 
beds, and in these flood prediction is very difficult. The em- 
banking of rivers, building of bridges, locks and other obstruc- 
tions may profoundly alter the flow not only locally, but for 
some distance up and downstream. Any narrowing or ob- 
struction raises the flood-level upstream by an amount which 
may exceed 10 feet and lowers it downstream. Low bridges are 
especially dangerous because they may stop ice, floating trees 
and other debris and so build up an effective dam. Local 
deepening of the river bed by dredging is of little use as the 
level is soon restored by the deposit of silt. 

The height of the greatest floods in the past can be found 
from natural marks on the banks, marks made on buildings or 
bridges to record the height, readings of flood gauges and the 
memory of the "oldest inhabitant." Natural marks are not 
always safe guides because for one thing capillary action may 



CLIMATIC ACCIDENTS 209 

raise the water level considerably inside porous stones, and for 
another, because they record the heights of wave-tops and not 
of the general flood level. A swift current piles up water against 
the piers of bridges and other obstructions and produces a local 
rise which does not extend to the banks. 

The memory of the "oldest inhabitant" is not to be neglected, 
when no other information is available. A warning is supplied by 
the building of a bridge in Scotland subsequently destroyed by 
flood. There were no written records, and the statements by local 
residents of the height to which the mountain stream had risen 
in the past seemed so incredible that the engineers dismissed them. 
Had they been believed the bridge might have been saved. 

The best basis for estimating the probable height of future 
floods is in actual records, either of the depth of water or of the 
volume of flow. Owing to the irregular shape of most river beds 
the height is not proportional to the volume, but by measuring 
a cross-section of the river bed the volume corresponding with 
any given level can be found; it is the sectional area of the 
river bed below that level multiplied by the average speed of 
flow. It has been found that the frequency of floods exceeding 
different volumes of flow in the Thames at Teddington in a 
given number of years is given by the expression 

log F= 19-4-4-88 log/ 

where F is the frequency per cent, and f is the flow in millions 
of gallons. There seems no reason why a similar rule (with 
different constants) should not apply to other rivers. Thus, 
if the highest flood level in a given period is known, this can 
be converted to the corresponding volume, and so to the 
highest level to be expected in another period of years. 
Moreover, by converting height to volume, the effect of 
narrowing or obstructing the river bed on the water level 
can also be calculated. The fact that narrowing the bed 
makes the river flow somewhat faster will probably not make 
a great deal of difference, but in any case can be written down as 
a safety margin. 

Measurements of rainfall are often available for long periods. 
If a record of river flow is available for a few years it may be 
possible to establish a relation between rainfall and river flow, 
from which the flow which probably followed the heaviest 
rainfall experienced can be calculated. 



210 CLIMATE IN EVERYDAY LIFE 

Floods can be minimised to some extent by providing easier 
run-off for the water (widening, deepening or straightening the 
channel), by building dykes or levees to control the water 
within bounds, or by spreading the peak of the flood over a 
longer period by constructing reservoirs to hold up the flood 
water. Afforestation of steep-sided valleys also serves to check 
the rate of run-off. Ground cultivated or covered by vegeta- 
tion can absorb from a half to two-thirds of the rain falling on 
it, where the same soil, bare and uncultivated, absorbs only a 
quarter or less. For a detailed study of the estimation of prob- 
able flood levels see B. D. Richards (1944). An account of the 
problem in more general terms is given by M. Parde (1946). 



HAIL 

Storms of large hailstones are very destructive, beating down 
crops, smashing glasshouses and windows, killing animals and 
even penetrating corrugated iron roofs. Their danger to wine 
crops is shown by the efforts made in France and Italy since 
early times to avert them, first by the ringing of church bells 
and later by shooting cannon at the thunder clouds, neither of 
which had any noticeable effect (see p. 260). There is also 
extensive insurance against loss by hail. According to J. B. 
Kincer (1937) the destruction of crops in the U.S.A. in the 
seventeen years 1909-25 amounted to 228 million bushels of 
wheat, 286 million bushels of oats and nearly 360 million 
bushels of corn. 

Hailstones vary in size from small, innocuous pellets of soft 
ice to solid spheres several inches in diameter. Since hailstones 
can only grow while being supported on violent up-currents of 
air they are mostly associated with severe thunderstorms. Com- 
parable statistics from different countries are hard to obtain 
because of great variation in size of hailstones and also of the 
very local distribution. H. Lemons (1942a) gives a map showing 
the distribution of hail in the U.S.A. in which the figures vary 
from less than two a year in the southern, eastern and north- 
eastern areas to six on parts of the Pacific coast and six to eight 
in Wyoming, Colorado and Nebraska. But the Pacific coast 
hailstones are small and soft while in the Middle West they are 
often large and destructive. At Potter, Nebraska, on 6th July 
1928, hailstones "as large as grapefruit" fell, one of which 



CLIMATIC ACCIDENTS 



211 



measured 15 inches in circumference and weighed ij lbs. Hail 
is not infrequent in Canada and does some damage to the 
tobacco, fruit and wheat crops. Severe hailstorms are frequent 
in the Transvaal, and hailstones weighing more than five 
ounces have been recorded, destroying tiled roofs and piercing 
galvanised iron. In the hailstorm of 25th December 1923 at 
Pretoria damage to property amounted to £80,000. Large 
hailstones have also been reported from Nigeria and Egypt. 
In Britain hailstones the size of "tennis balls" have been re- 
corded, killing chickens and piercing corrugated iron and the 
fabric roofs of motor-cars. On nth May 1945 one hailstone, 
"not the largest," was found to weigh 8J ozs. Judging by the 
holes made in an asbestos roof, even larger stones may have 
fallen in Northamptonshire in September 1935. 

According to E. G. Bilham and E. F. Relf (1937), the limiting 
velocity of fall of spherical hailstones increases abruptly when 
the weight of the hailstone reaches if lbs., and that weight is 
therefore about the maximum possible. The plains of India 
suffer from storms of large hailstones, descriptions of which were 
collected by J. Eliot (1899) and a statistical analyses of these 
data supported Bilham's conclusion (G. E. P. Brooks, 1944). 

The terminal velocity of a hailstone in still air depends on its 
size and density (i.e. on the amount of air included in the ice) . 
For a specific gravity of o-6 Bilham and Relf give the following 
theoretical velocities : — 



Table 25. — Terminal velocity of hailstones. 



Diameter, inches 


1 
i 


i 


I 


2 


3 


4 


5 


Weight, ozs. . 
Velocity, feet/second 
m.p.h. 


0-003 

29 
20 


0-02 

42 
29 


0-18 

59 
40 


i*5 

84 

57 


4-9 
104 

7i 


1 1 -6 

124 

9i 


22 * 7 

323 
220 



When the diameter exceeds about 4J inches, the theoretical 
velocity is much higher (323 ft. /sec, or 220 m.p.h. with a 
diameter of 5 inches), but Bilham considers that it is very 
doubtful whether this is ever reached in nature. But even a 
mass of 12 ozs. hurtling through the air at 90 m.p.h. can do 
considerable damage. 

At places in the British Isles hail is recorded from three up 
to more than twenty times a year (London, Met. Office, 1923). 



212 CLIMATE IN EVERYDAY LIFE 

The larger frequencies are found in the west, but they refer 
almost entirely to unimportant falls of small or soft hail. Damage 
by hail is mostly confined to the east and Midlands, destructive 
storms beating down crops and breaking glass and tiles occur- 
ring at intervals of a few years. At any one place the frequency 
is of course less than this, and individual farms may escape 
altogether for decades. It is recorded that the storms of 24th 
June 1897 in Middlesex and Essex, in which hailstones "as 
large as hens' eggs" fell with violent winds, caused great 
distress among farmers as there had not been a bad hailstorm 
for some years and many of them had ceased to insure against 
hail. Close to the Equator hail is less frequent because even 
at heights of 20,000 feet or so the air is too warm for its forma- 
tion. According to H. Lemons (1942^) hail is small and rare 
in tropical islands, and in the tropics generally it is more fre- 
quent and severe at high altitudes than in the lowlands. In 
high latitudes there is probably no true hail. 



SNOW 

Heavy or continuous snow interferes with rail and road 
traffic. In Britain snow falls on a few days a year in the south- 
west, ten to fifteen days in the south-east, increasing to twenty 
to twenty-five days in the Midlands and forty to fifty days in 
the Highlands of Scotland. In colder parts of the world snow 
is very frequent in winter. Many falls of snow, however, are 
small and either melt rapidly or add little to the existing snow 
cover ; heavy falls are rare and occur in only a sm all proportion 
of the years. Accounts of some of the worst storms in recent 
years in Britain are given in Chapter III (p. 100) and in the 
U.S.A. in Chapter IV (p. 120). For a discussion of the pro- 
tection of railways from drifting snow by fences or plantations, 
and clearance of drifts, see W. K. Wallace (1949). 

In countries with cold winters a persistent snow cover forms 
every year. A map of the average duration of snow cover is 
given in Fig. 8, but there are wide variations from year to 
year. 

Snow on skylights interferes with the top-lighting of rooms by 
daylight. The loss of daylight is mainly due to the high re- 
flectivity of a snow surface ; even a very thin layer of new snow 
reflects about 80 per cent, of the light falling on it. From some 



CLIMATIC ACCIDENTS 213 

observations by L. C. Porter (1934) in Cleveland, U.S.A., it 
appears that half an inch of light fluffy snow further reduces 
the illumination to about 7! per cent., 1 inch to 3 per cent., and 
3 inches to only 1 per cent, of that above the snow. In the case 
observed by Porter, an outside illumination of 247 foot-candles 
was reduced to 2-foot candles by 3 J inches of snow on a skylight. 

Glazed frost ("ice storm") is described on pp. 47 and 120. 
Widespread glazed frost is fortunately very rare in Britain, 
occurring only at intervals of several years ; when it does occur 
it completely dislocates road and even rail transport, and often 
telephonic and telegraphic communication as well. The 
damage which probably takes longest to repair is the wholesale 
breaking of telephone and telegraph wires and even the snap- 
ping or uprooting of whole rows of poles. 

Glazed frost is also a source of trouble on overhead trans- 
mission lines of electric power systems. According to H. W. 
Grimmitt (1945) the most frequent cause of damage is short- 
circuiting and burning of the insulation. The conductors are 
weighed down by the ice, and when it falls off one conductor 
this rises abruptly and may come in contact with higher con- 
ductors. The conductors and even the supports may break 
under the weight of ice, especially when the wind is strong. 
Strong winds are rare during the actual formation of glazed 
frost, but may spring up afterwards, while the ice is still on the 
wires. Grimmitt gives a tentative map of the distribution of 
"ice-storms" which affect overhead transmission lines. This 
shows frequent occurrences in the hill country of the Pennines 
and South Wales, where above 1,000 feet trouble may be ex- 
pected every other winter, "infrequent" damage on the 
eastern side of the Pennines, in North Wales, parts of the Mid- 
lands, Gloucestershire, northern Devon and Somerset, Wiltshire 
and Hampshire, and none elsewhere except for a small area on 
the North Downs of Kent. Of the twenty-eight years, 1916-43, 
damage was reported in thirteen years, but in five of these it 
was only local and slight. 

LIGHTNING 

A lightning flash consists of a series of discharges of electricity 
following one another very quickly along the same track. 
Most flashes travel from one point in the cloud to another; 



214 CLIMATE IN EVERYDAY LIFE 

some travel from cloud to earth and possibly a few from earth 
to cloud. A lightning flash is usually branched, the branches 
pointing in the direction of travel, so that a discharge from 
cloud to earth usually reaches the ground along several 
channels. Near the ground a discharge from earth to cloud is 
concentrated in a single channel and is generally more intense. 
B. F.J. Schonland (1938) doubts the existence of earth-cloud 
flashes, but G. C. Simpson (1929) shows a photograph of 
flashes branching upwards. These papers by Simpson and 
Schonland give detailed descriptions of the complicated struc- 
ture of lightning. 

Because of its spectacular and terrifying nature lightning has 
always played a large part in popular imagination, but though 
the total annual loss of life and property from lightning over 
the world is large it is probably much smaller than that due 
to other climatic accidents such as floods and hurricanes. The 
effects of lightning may be considered under: destruction of 
life ; destruction of property mainly by fire ; interference with 
electric power systems. 

The number of persons killed by lightning depends partly on 
the frequency of thunderstorms (see below) and partly on the 
local conditions. For example, the risk of death by lightning 
in a town is very much smaller than in the country, and it is 
greatest in hilly or sparsely wooded country. Most of the deaths 
occur among men or boys sheltering from rain under trees or 
in small outbuildings. Cattle and sheep are also killed under 
trees and near wire fences. The most complete statistics avail- 
able were provided by the Metropolitan Insurance Go. (New 
York, 1948). These give the annual death-rate in the U.S.A. 
as about 400 or 3 per million. Nine-tenths of these occurred 
in places with 2,500 or fewer inhabitants, though such places 
contain only 40 per cent, of the total population. The death- 
rate was greatest in the Middle West and Gulf States, exceeding 
six per million in Montana, Idaho, Wyoming, North and South 
Dakota, Colorado, New Mexico, Arkansas, Mississippi, Ala- 
bama and Florida. It was under one per million in the Pacific 
coast States of Washington, Oregon and California, where 
thunderstorms are few and mild, and in the largely urbanised 
north-eastern States of New Jersey, New York, Rhode Island, 
Massachusetts and New Hampshire. In Great Britain the 
death-rate is probably well under one per million. 



CLIMATIC ACCIDENTS 215 

Lightning causes some structural damage to buildings and 
sometimes kills trees, but by far the greater portion of damage 
is due to fires initiated by lightning, especially forest fires. 
J. B. Kincer (1937) estimated the loss in the U.S.A. from fires 
caused by lightning as 12,000,000 dollars a year. A very costly 
fire was started by lightning striking a large oil tank. In forest 
regions the risk is greatest when thunderstorms follow a long 
period of fine weather during which the surface of vegetable 
debris becomes very dry and tinder-like. Warnings of such 
conditions are given by the U.S. Weather Bureau and a watch 
is kept. Lightning striking the wires of an overhead electric 
transmission service causes sudden surges of current which 
break down the insulation of the wires and sometimes cause 
transformer failures. Protection from lightning is discussed in 
Chapter XII. 

Lightning flashes between one part of the cloud and another 
do no damage, and we are interested only in the frequency of 
cloud-earth flashes. Adequate statistics of these are available 
for only a very few places. R. H. Golde (1945) quotes various 
estimates which range from 1-5 to 9-5 flashes per square mile 
per annum. He estimates the most probable figure for Britain 
as six cloud-earth flashes per square mile per annum. The 
average annual number of thunderstorms ("isokeraunic level") 
is about twelve, and this gives us the rough rule that number 
of cloud-earth flashes per square mile is half number of thunder- 
storms in the vicinity. The same rule is adopted in the U.S.A. 
(R. H. Golde, 1946). 

The ratio must also depend on the severity of the thunder- 
storms. These are most severe over inland areas, and on the 
north-west and west coasts of Europe and North America they 
are relatively mild and innocuous. 

In the tropics thunderstorms give the impression of being 
more severe, but owing to the greater height of the clouds a much 
greater proportion of the flashes are cloud to cloud and cor- 
respondingly fewer are cloud to earth. Exact figures are not 
available, but the percentage of flashes reaching the earth in 
tropical regions may be a half or less of that in temperate regions. 
J. A. Chalmers (1941) shows that the higher the dew point of 
the air (i.e. the temperature at which water is condensed out of 
the atmosphere), the more probable is a lightning flash within 
the cloud in comparison with one to earth, the ratio being 



2l6 CLIMATE IN EVERYDAY LIFE 

roughly proportional to the dew point in ° C. On the other 
hand, the total number of flashes per storm is greater in tropical 
than in temperate storms. Nevertheless, one hears compara- 
tively little about thunderstorm damage in tropical regions, and 
provisionally the number of cloud-earth flashes per square mile 
in the tropics may be taken as one-third of the number of 
thunderstorms. 

The frequency of cloud-earth flashes probably increases with 
the height of the ground, but it depends very much on the 
nature of the surface. Towers are weirknown to be struck more 
frequently than surrounding flat ground. In U.S.A., with 
twenty-five to forty-five thunderstorms a year, the number of 
strokes on buildings up to 500 feet high increases linearly at the 
rate of 0-004 H P er annum, where H is the height in feet. Above 
500 feet the frequency increases more rapidly, and the Empire 
State Building, 1,250 feet high, was struck sixty-eight times in 
three years (New York, Inst. Radio Engrs., 1943). Mountains 
high enough to extend into the thunderclouds are very fre- 
quently struck, but the voltage of the strokes is small and they 
are less dangerous than flashes to low ground. According to 
G. F. Brooks (1935) all the superstructures, radio electric 
recording instruments and telephones of Mount Washington 
Observatory flicker with brush discharge during thunder- 
storms, but although the slopes are bare there are no records 
of death by lightning. 

The annual frequency of thunderstorms was mapped by 
C. E. P. Brooks (1925). On the basis of this chart and of the 
data given above a sketch map of the probable frequency of 
lightning flashes to earth has been constructed (Fig. 27). This 
gives the estimated number of flashes per square mile per year, 
but in any locality there are probably local variations due 
to nature and elevation of ground, vegetation, buildings, 
etc. The causes of such variations are obscure: D. Muller- 
Hillebrand (1937) states that there is doubtful evidence that 
disturbances of conductors are most frequent where the ion- 
content of the air is large. It has been suspected, but probably 
on insufficient evidence, that underground water attracts 
lightning. The main point is to ensure that the masts are 
properly earthed and this may require special precautions in 
dry regions. 



CLIMATIC ACCIDENTS 



217 




u 

cr 

a, 

2 






■a 



be 



2l8 CLIMATE IN EVERYDAY LIFE 



TORNADOES AND SQUALLS 



The true tornado is a whirling column of air with an average 
diameter of about 1,000 feet. The wind reaches very high 
speeds; these have never been accurately measured, since no 
instrument can stand up to them, but estimates from the weight 
of objects lifted put the maximum speed in a severe tornado as 
over 300 m.p.h., and sometimes as high as 500 m.p.h. The 
winds often have an upward component near the centre which 
has been estimated to reach 100 to 200 m.p.h., so that large, 
heavy objects may be carried up to heights of several hundred 
feet and transported for considerable distances. The centrifugal 
force of the rotating winds causes a partial vacuum in the centre. 
The damage is of two kinds ; the wind blows or flattens down 
houses and trees and carries away cars and other solid objects 
in its path, and the sudden reduction of pressure causes build- 
ings to "explode". The air inside a building is not given time 
to adjust its pressure to that of the centre of the tornado, 
especially if doors and windows are closed, and there is an 
outward pressure on walls, windows and roofs which may 
amount to two or three pounds to the square inch. Walls and 
windows are blown outwards and roofs are lifted off. This 
suction effect is responsible for most of the "freak" effects of 
tornadoes, such as plucking chickens and stripping people of 
their clothes. Only the strongest and sturdiest buildings can 
withstand the combined effects of wind force and suction. In 
the tornado of 1936 at Gainesville, Ga., however, buildings 
with strong steel frameworks suffered only minor damage, such 
as shattered windows, while lighter structures were completely 
destroyed. 

When a tornado strikes a town the damage is very great. The 
outstanding example is the St. Louis tornado of 27th May 1896, 
which swept through the town and destroyed property worth 
10,000,000 dollars; 306 people were killed. J. B. Kincer (1937) 
states that in the years 1916-35 2,800 tornadoes were reported 
in the U.S.A., costing the lives of 5,224 people and damage to 
property amounting to 230,000,000 dollars. C. W. Brown and 
W. O. J. Roberts (1937) estimate the direct damage to pro- 
perty as 250,000,000 dollars in fifty years. A single tornado on 
the average devastates an area of little more than one square 
mile, so that the total area affected in twenty years was only 



CLIMATIC ACCIDENTS 



219 




220 CLIMATE IN EVERYDAY LIFE 

about 3,000 square miles. Since the area of the U.S.A. subject 
to tornadoes is about 1,250,000 square miles the odds against 
a tornado striking any one place in any one year are at least 
10,000 to one. 

Tornadoes occur most frequently in the great lowland areas 
of the central and upper Mississippi, Ohio and lower Missouri 
valleys. They also occur in Georgia, North and South Carolina 
and the inland parts of the Gulf States ; they are rare on the 
coasts and practically unknown west of ioo°W. long. The 
States from which records are most frequent are : Kansas, Iowa, 
Texas, Arkansas, Illinois and Missouri. A generalised picture 
of the distribution is shown by the vertical shading in Fig. 28. 
A more detailed study has been made by C. W. Brown and 
W. O. J. Roberts (1937). They take as their unit one mile of 
tornado track one-tenth of a mile wide and divide the number 
of units observed in a county in fifty-one years into the area of 
the county in square miles. The answer, multiplied by 510, 
gives the odds against a tornado at any place in that county 
in any year. The figures are very irregular, partly because in 
the less densely peopled areas many tornadoes may go unre- 
ported, but possibly also because there may be real local differ- 
ences in the frequency of tornadoes. The smallest value on 
Brown and Roberts' map is about five, indicating odds of about 
2,600 to 1 against a tornado in any one year even in the worst 
areas. The risk is minute and is hardly worth bothering about. 
According to C. F. Brooks (1935) the insurance rate against 
damage by wind-storm in the eastern and middle States re- 
presents an expectation of total destruction once in over 1,200 
years for dwellings or once in over thirty years for more flimsy 
structures, but this includes insurance against any strong wind. 
Many houses in the areas most affected are provided with 
strong cellars into which the occupants can retire on the ap- 
proach of a tornado. T. A. Blair (1937) considers that the 
"cyclone cellar" should be completely underground and away 
from any other building. He adds that an automobile can 
generally outrun a tornado by travelling in a direction at right 
angles and to the left of the direction in which the tornado is 
advancing (i.e. if the tornado is moving north-east go towards 
the north-west). 

Tornadoes of the American type occasionally occur in other 
parts of the temperate regions, but are less frequent and 



CLIMATIC ACCIDENTS 221 

generally less severe. In Britain the most notable was the 
South Wales tornado of 27th October 19 13 which caused a 
great deal of damage along a sharply bounded track several 
hundred feet wide ; trees were uprooted and buildings destroyed 
(H. Billett, 1 9 14). Other examples were a tornado and hail- 
storm in Derbyshire on 12th May 181 1, Dublin on 18th April 
1850, Cowes on 28th September 1876, Whipsnade on 18th 
January 1945, and Birmingham on 4th February 1946, besides 
a number of minor ones which did only very local damage. 
These whirlwinds are generally smaller, less violent, and 
shorter-lived than the American variety, and though several 
probably occur each year somewhere in Britain the risk to any 
particular building is negligible. Similar visitations are some- 
times reported from other parts of the world, but they are rare 
near the Equator. For an account of tornadoes in Europe see 
A. Wegener (191 7). 

Squalls differ from tornadoes or whirlwinds in the absence of 
rotary motion about a vertical axis. A squall is a sudden rise of 
wind velocity, often with a marked change of direction. Squalls 
are of three types: — 

(1) A mass of cold air advancing along a broad front, under- 
cutting and violently lifting the warmer air in front of it. There 
is generally a brief but heavy shower of rain or hail, frequently 
with thunder, and a sudden drop of temperature. This type is 
known as a "line-squall," and is commonly associated with the 
passage of the "cold front" of a depression. This type of squall 
is most severe in middle latitudes, but it can occur also in the 
weaker depressions of sub-tropical regions. In Australia it is 
known as the "southerly burster," in the Argentine as the 
"Pampero" and in the southern Mediterranean as a "white 
squall," so-called because it brings no cloud or rain. The greatest 
damage is usually patchy, in strips about 30 by 100-150 yards 
(H. Faust, 1948). The so-called "tornadoes" of West Africa are 
also of this type, though local rotary motion sometimes develops. 

These squalls are sometimes violent enough to do minor 
damage on land, but they are more dangerous to shipping 
because of the suddenness with which they spring up and the 
rapid change of wind direction. On 24th March 1878 the 
training ship Eurydice foundered during a line-squall. Because 
of their powerful vertical currents and great lateral extent 
line-squalls are dangerous to aircraft. 



222 CLIMATE IN EVERYDAY LIFE 

(2) Somewhat similar are the squall lines associated with the 
descent of cold air after passing over mountain ranges. The air 
often develops a violent rotation about a horizontal axis. Such 
squalls are met with in many parts of the world, but are 
especially frequent in the tropics; the Sumatras of the Malacca 
Strait are among the best-known examples. These squalls rarely 
do much damage on land, but are troublesome to shipping. 

(3) Thunderstorms are sometimes accompanied by local 
squalls. The centre of a thunderstorm is a region of violent 
uprush of air, especially where hail is produced. This air is 
cooled and descends on the edges of the storm, mostly in front 
of it, where the contrast of temperature between the descending 
air and the undisturbed warm surface air is greatest. The down- 
rush is increased by the heavy rain or hail which sweeps the air 
along with it. The resulting squall comes up suddenly and is often 
strong enough to do minor damage, but the effects are usually 
local, limited to an area up to 5 miles long and a half mile wide. 

In an example at Kano in Northern Nigeria on 12th June 
1947, described by A. T. Dorrell (1947), the wind rose sud- 
denly from 20-80 m.p.h. along a front 200-300 yards wide, and 
several buildings were unroofed. At the same time temperature 
fell instantaneously from 89°-66° F., and there was a short 
burst of heavy rain. Thunderstorm squalls are most frequent 
and severe in the transition period preceding the rainy season 
in sub-tropical countries. 

In addition the ordinary cyclonic depressions of temperate 
latitudes and the cyclones, typhoons or hurricanes of tropical 
and sub-tropical regions (p. 225) include in their general wind 
circulation patches of stronger winds which to some extent 
resemble squalls, but are of longer duration. In temperate 
regions such winds are often associated with the passage of 
secondary depressions, and form narrow belts. A notable 
example occurred on 24th March 1895. A deep depression was 
centred over the Shetlands and a small, intense secondary 
depression traversed the Midlands. On the south-east of this 
secondary the winds exceeded gale force over a track 30-50 
miles wide where a great deal of damage was done to buildings, 
many trees were uprooted and several people were killed. One 
of the minor troubles caused by squalls is the blocking of roads 
by trees blown across them; in severe storms this may result in 
a good deal of interruption to traffic. 



CLIMATIC ACCIDENTS 2^3 

An attempt has been made to estimate the maximum gust 
velocities to be expected at a height of 33 feet, once in ten years, 
in different parts of the world. Observations with gust-recording 
anemometers are rare except in Britain, parts of North America, 
India and a few places like Hong Kong and Singapore. Gup 
anemometers are more plentiful; these give the average wind 
(mean of gusts and lulls) which can be converted to gust 
velocity by using an estimate of the "gustiness factor" (p. 86). 
For many parts of the world only estimates based on personal 
impressions or accounts of damage are available. The details 
of the measurements or estimates are available in the Meteoro- 
logical Office, London; they may be summarised as follows: — 

Exceeding 125 m.p.h. 
West coast of Iceland. 
Part of the West Indies, including Porto Rico, Haiti-San Domingo, 

Jamaica and the eastern half of Cuba. 
Coast of China around Hong Kong. 
Northern end of Formosa. 
Very exposed positions in eastern U.S.A. 

Exceeding 100 m.p.h. 

South and east coast of Greenland and most of Iceland. 

North and west coasts of Norway as far south as 6o° N. 

West coasts of Scotland, Ireland, Pembroke, extreme south-west 

of Cornwall, Scilly Islands. 
Exposed positions in interior of U.S.A., from North Dakota and 

Wisconsin in the north to New Mexico-Missouri in the south, 

and from Colorado in the west to Illinois in the east. 
Coast of Florida, Georgia and Carolina. 
District round Rangoon (Burma). 
Coast of China from Hainan to southern tip of Japan. 
Southern Sakhalin and Hokkaido. 
A small coastal strip of north-west Australia between Broome and 

Wyndham. 
South-west coast of Chile south of 50 S. 

Exceeding 80 m.p.h. 
Aleutian Islands. 

Most of U.S.A. except valley regions in Rocky Mountain system 
and area between Alleghenies and a short distance inland from 
Atlantic coast. 
Coast of Gulf of Mexico and all West Indies. 

Coast of West Greenland to 70 N. and of East Greenland to 75 N. 
The whole of the British Isles. 



224 CLIMATE IN EVERYDAY LIFE 

Norway, except where sheltered from the west. 

Arctic coast of Finland and Russia. 

Coast of Baltic (except Gulf of Finland), all Denmark, coastal 

regions of north-west Germany, Holland, Belgium. 
France north of La Rochelle. 
West coasts of Corsica and Sardinia, Malta and north coast of 

Africa from Tripoli to Morocco. 
Coast of Palestine. 

Bay of Bengal and neighbouring regions. 
Coast of Indo-China and China from about io°-30° N. 
Southern Japan and Korea. 
North-west Australia from 20° N. to Darwin. 
East coast of Australia from Brisbane to Sydney and coasts of 

Tasmania. 
Islands of Mauritius, New Hebrides, Fiji, etc. 
East coast of South America from Porto Alegre to Bahia Blanca. 
Coast of Chile south of 45 S., Tierra del Fuego. 

Over most of the remaining parts of the world the gust velocity 
to be expected is about 60 m.p.h. The lowest values (below 
45 m.p.h.) are found in the neighbourhood of the Equator — 
South America between io° N. and io° S., Central Africa 
from 5 N.-20 S., northern India under the shadow of the 
Himalayas, and the East Indies between about 5 N. and S. 

These figures represent average open conditions. Very ex- 
posed sites on hill-tops and headlands naturally experience 
higher gusts; on the other hand, there are plenty of more 
sheltered situations, such as valleys transverse to the strongest 
winds, lee slopes of hills, and forest clearings, where the winds 
are less strong. 

The highest velocity to be expected in a given period naturally 
depends on the length of the period, but to a much less extent 
than would be expected. In a single year chosen at random the 
highest gust to be expected is about 80 per cent, of the ten-year 
maximum. The hundred-year maximum is about 120 per cent, 
of the ten-year maximum. The maximum velocity V N to be 
expected in JV years may be related to the maximum V 10 in 
ten years by the approximate expression 

V N =V 1Q (o-8+o-2 log Jf). 

The gust velocity increases with height above the ground, but 
not so rapidly as the mean velocity. An expression developed 



CLIMATIC ACCIDENTS 225 

by N. Carruthers (1943) gives the following average ratios 
between the gust velocity at a height H and that at 33 feet: — 



Height, feet . 


10 


20 


33 


50 


60 


80 


100 


VIVS3 

Coast 
Inland 


0-93 
0-97 


0-97 
0-98 


i-oo 

I -00 


1-03 

1-02 


1-06 
1 03 


1-07 
1*04 


1-09 
1-05 



The "roughness" of the ground affects the vertical variation of 
gust velocity. If the ground to windward of a building is 
occupied by shrubs, small trees, fences and similar obstacles, 
the speed near the ground is decreased, but that at roof height 
of average buildings may be increased compared with more 
open situations. 



The "tropical cyclone" is found between about 5 and 30 
latitude, on the western sides of the Pacific, Indian and North 
Atlantic Oceans, in the Bay of Bengal and off north-west 
Australia. It does not occur in the South Atlantic. The winds 
blow round, and slightly towards, a central area of low pres- 
sure (the "eye" of the storm), and may reach speeds up to 180 
m.p.h. At San Juan, Porto Rico, in 1928 the wind had reached 
this speed when the anemometer blew away, and it continued 
to increase in strength for some time afterwards. Winds of 
125-140 m.p.h. have been recorded in a number of hurricanes 
and typhoons. The passage of a tropical cyclone is accompanied 
by heavy rain and very rough seas. In the northern hemisphere 
the winds blow round the centre in the opposite direction to 
the hands of a clock (anti-clockwise rotation). On the right 
of the track along which the cyclone is moving the speed at 
which the centre is advancing is added to the speed of rotation, 
so that the strongest winds are found in this sector. These winds 
drive before them a great mass of water, which strikes the coast 
as a storm wave, and may be 40 feet above the general level of 
the sea. When the storm wave comes at high tide it often does 
more damage than the wind and rain. In the southern hemi- 
sphere the winds blow clockwise about the centre and are 
strongest to the left of the track. 

The regions in which tropical cyclones are found are indicated 
by horizontal shading in Fig. 28. Light shading shows regions 



226 



CLIMATE IN EVERYDAY LIFE 



where visitations are rare (at any individual place only once in 
many decades) ; heavy shading indicates more frequent storms. 
Cyclones have a variety of names. In the Arabian Sea, Bay of 
Bengal and South Indian Ocean they are known as "cyclones," 
and in north-west Australia they are "willy-willies." In the 
West Indies, Gulf of Mexico and coast of Florida, and in the 
South Pacific they are termed "hurricanes." In the China 
Sea they are "typhoons" and in the Philippines "baguios," 
but all these names refer to the same phenomenon. 

Cyclones generally originate over the ocean in about latitude 
5-1 o° N. or S. At first they move eastwards with a trend away 
from the Equator, which increases with increasing latitude. In 
a latitude which varies with the season from i5°-25° they 
"recurve" and turn westwards, travelling north-west in the 
northern hemisphere and south-west in the southern hemi- 
sphere. The general lines of the tracks are shown by the shaded 
areas of Fig. 28. Some cyclones, however, follow unusual 
tracks, and they may not recurve at all. Eventually a cyclone 
either strikes land (when it usually breaks up quickly with 
torrential rain), dies out over the ocean, or passes into tem- 
perate latitudes, where it loses the character of a tropical storm. 

Statistics of tropical storms are not complete, as some storms 
have undoubtedly passed unrecorded, especially in the South 
Pacific. The following table shows the number observed in 
each month, corrected to a period of fifty years to make the 
different regions comparable. 



Table 26. — Number of Tropical Cyclones in Fifty Years. 





Jan. 


Feb. 


Mar. 


April 


May 


June 


July 


Aug. 


Sept. 


Oct. 


Nov. 


Dec. 


Year 


West Indies 












15 


18 


47 


82 


66 


11 


3 


242 


Arabian Sea 


1 


— 


— 


2 


6 


7 


2 




1 


7 


8 


2 


36 


Bay of Bengal . 


— 


— 


— 


6 


12 


3 


5 


1 


8 


14 


15 


6 


70 


China Sea and 




























Western N. 




























Pacific . 


59 


33 


35 


27 


63 


65 


173 


177 


212 


183 


100 


65 


1192 


Eastern North 




























Pacific . 


— 


— 


— 


— 


3 


29 


37 


39 


89 


45 


5 


3 


250 


S. Indian Ocean 


66 


84 


61 


28 


9 




— 


— 




3 


12 


39 


302 


N. W. Australia 


11 


7 


8 


3 




— 


i 


— 


i 


fr 


1 


5 


35* 


Queensland 
South Pacific 
i6o°E.-i40°W. 


29 


23 


27 


12 


5 


7 


6 


— 


4 


3 


2 


8 


126 


22 


16 


21 


6 


1 


1 


I 


I 


1 


1 


2 


11 


83 



The figures for the South Pacific are based on records for the 
period 1 789-1 924, and are certainly too low. 



CLIMATIC ACCIDENTS 227 

Only about half or one-third of these storms are classed as 
severe; some of them bring nothing worse than strong winds. 
The tracks of the majority lie entirely over the oceans, and 
affect only a few small islands at most. For example, in the 
West Indies the average annual number of hurricanes reported 
is nearly five, but the average frequency of destructive hurri- 
canes is less than one a year. 

When a severe storm does strike a populous coast or island 
the damage is very great. J. B. Kincer (1937) estimated the 
damage by hurricanes in the United States during the twenty 
years 1916-35 as very nearly 386,000,000 dollars. I. R. Tanne- 
hill (1944) gives the damage to property in the United States 
during the period 191 7-41 as about 500,000,000 dollars, but 
states that more than half of this occurred in the New England 
hurricane of September 1938. He puts the loss of human life 
during the same period as about 4,200 in the United States, but 
above 10,000 if the West Indies, Central America and Mexico 
are included. 

De Monts (1936) considers that the cyclones of the South 
Indian Ocean are on the average much less severe than those 
of the West Indies region, the maximum wind speed averag- 
ing only two-thirds as great. In the south-eastern U.S.A. 
the highest waves are 30 feet above the general sea-level, in 
Reunion Island (21 S., 56 E.) rarely as much as 15 feet. Since 
the destructive power of a wave is proportional to the square 
of its height, the damage by waves is only a quarter as great 
at Reunion as in the U.S.A. On the other hand, the cyclone 
rainfall is generally greater in the South Indian Ocean than 
in the Gulf States. References to some of the worst disasters 
in the West Indies, south-eastern U.S.A., India, China and 
Japan are given on pages 124, 127, 133 and 137. 

The diameter of a tropical cyclone is small near its origin, 
averaging 25-50 miles, but as it moves into higher latitudes it 
expands to an average diameter of 500 miles. The central ring 
of hurricane winds surrounding the "eye" is much smaller, 
however, probably averaging less than 100 miles. According 
to the Admiralty Weather Manual (London, 1938, p. 376) the 
average velocity 35 miles from the centre exceeds 75 m.p.h.; 
50 miles from the centre it is 70 m.p.h., decreasing to 30 m.p.h. 
at 150-200 miles. For more severe cyclones these figures must 
be increased proportionally. The average width of the belt of 



228 CLIMATE IN EVERYDAY LIFE 

destruction is about ioo miles. The hurricane tracks in a 
region such as the West Indies are spread over a belt which has 
a width of the order of 1,500 miles, so that any cyclone will bring 
hurricane winds over less than one-fifteenth of the belt. More- 
over, all cyclones do not traverse the whole length of the belt ; 
some form and others die out within the region. Thus an 
average frequency of five cyclones a year in a belt 1,500 miles 
wide is equivalent to only one in about five years at any par- 
ticular spot. Further, only about one in three of these cyclones 
is severe, so that the incidence of catastrophic storms is further 
reduced to one in about fifteen years. The risk is greatest in the 
centre of the heavily shaded belts and decreases towards the 
edges. 

From some data given by I. R. Tannehill (1944) for the 
years 1879- 1943, the average interval in years between storms 
of hurricane intensity per 100 miles of coast, in the south- 
eastern United States, is found to be : — 



Mississippi 


. 4 


North Carolina 


9 


Alabama . 


5 


South Carolina 


12 


Texas 


7 


Louisiana 


14 


Georgia . 


. 8 


Florida . 


17 



The Meteorological Services of the U.S.A., India, Philippines, 
etc., follow the tracks of cyclones and broadcast warnings of 
their approach. These warnings have saved many lives, both 
on land and at sea. 

DUST STORMS 

One of the worst features of strong winds in arid and semi- 
arid regions, especially during dry periods, is the dust. In the 
dry years 1934 and 1935 the dust nuisance in the south-central 
plains of the U.S.A. became so great that this region, including 
parts of Colorado, Kansas, New Mexico and Oklahoma was 
christened the "Dust Bowl." On 1 ith May 1934 the dust cloud 
was 900 miles wide and 1,500 long; it darkened the sky over 
New York and Boston, while farther west snow-ploughs had 
to be used to clear the highways of drifted soil (G. F. Brooks, 
1935). The dust lowers visibility to a few feet, interferes with 
road and rail transport, and abrades all surfaces against which 
it is blown. Automobiles travelling through a dust storm have 
been stripped of paint. The removal of the top soil has a 
disastrous effect on the fertility of the ground; in some dust 



CLIMATIC ACCIDENTS 2 29 

areas even weeds cannot grow and the ground is rapidly eroded. 
Dust storms are now regarded so seriously in the U.S.A. that 
they are reported annually in the Monthly Weather Review. 

Dust storms also occur in Europe, but are only serious in the 
south-east, in southern Russia in particular. One such dust 
storm occurred on 26th-2gth April 1928, when artificial light 
had to be used all day in eastern and central Europe, and in 
southern Russia the dust formed heaps like snowdrifts more than 
a foot deep. The worst dust storms occur in desert regions, and 
are typified by the Haboobs of the Soudan (see p. 161). 

When rain falls through a dust cloud it comes down as mud. 
The worst mud-rains follow or accompany volcanic eruptions, 
but even surface dust can produce unpleasant results. When 
the dust originates in a desert it is frequently red, and dust from 
the Sahara is the basis of the occasional "blood rains" of 
Europe. A notable red rain fell over the southern half of 
England and Wales and a large part of Europe on 2ist-23rd 
February 1903. The origin of the dust was traced back to the 
Sahara. Mud and sand rains make stains which are difficult 
to remove, but are otherwise harmless. 

Most of the phenomena described in this chapter are on so 
large a scale, or draw on such a great store of natural energy 
that man can do little against them. Attempts at breaking 
up hailstorms are described in Chapter XII, but their success 
is very doubtful. The only protection against hurricane winds 
and tornadoes is by solid building. Protective systems against 
lightning (see p. 262) are good but not infallible. Probably man's 
most successful counter-measures against weather are special 
weather forecasts — warnings of heavy rain, flood, gale, thunder, 
snow, hail, frost, which enable precautionary measures to be 
taken in good time. The risk of loss by any of these climatic 
"accidents" is slight at any one spot in any one year, but when 
loss does occur it is often very heavy. The risk can be shared by 
insurance which is largely developed, especially in the U.S.A. 



PART III 
THE CONTROL OF CLIMATE 



CHAPTER XI 

HEATING, AIR CONDITIONING, LIGHTING, 
CLOTHING 

A LARGE part of both the working time and the leisure 
of many people is spent indoors, and in order that they 
k may maintain their efficiency the climate of the buildings 
in which they live, work or play must be kept within certain 
limits of temperature, humidity and illumination. The outdoor 
climate of most places, however, is outside these limits at some 
seasons, and the climate inside buildings must be controlled. 
This control takes three forms: — 

( i ) Heating of rooms in cold weather. 

(2) Air conditioning, or control of humidity as well as tem- 

perature, generally by cooling and drying the air. 

(3) Artificial lighting in dull weather and at night. 

Out of doors protection is given by suitable clothing. The pur- 
pose of clothing varies according to the weather; apart from 
adornment and carrying capacity (with which we are not con- 
cerned), we have to consider: — 



(1) Protection against cold. 

(2) Protection against rain. 

(3) Protection against insolation. 



HEATING 

The temperature which is regarded as most comfortable for 
sedentary work is 60-65 F. in Britain and 65-70 F. in North 
America. This standard temperature varies somewhat accord- 
ing to the nature of the activity carried on; in the U.S.A., for 
example, it varies from 61 ° F. for occupations which involve a 
good deal of bustle to 69 F. in residential apartments, but for 
practical purposes the figures quoted above may be taken as 
good averages. 

The outside air has a diurnal range of temperature, the mid- 
day hours being from 10-20 F. or more warmer than the early 



234 CLIMATE IN EVERYDAY LIFE 

morning hours, but the structure and contents of most buildings 
have a considerable power of storing heat, and the day tem- 
perature of unheated buildings changes much less from day to 
night than that of the outside air. The range depends on the 
size of the building, the thickness of the walls, and the volume 
of contents. Large unpolished metal objects such as machinery, 
and tanks of water, absorb a good deal of heat during the day 
and give it out again by night, and so tend to stabilise the tem- 
perature. R. Grierson (1941) points out that the nature of the 
walls is important; rooms with panelled walls warm up and 
cool down much more rapidly than rooms with walls of stone 
or brick. Further, human beings, cooking apparatus, lights, 
etc., produce a good deal of heat additional to that supplied by 
the heating installation. Experience has shown that the amount 
of power required for heating in any building is proportional to 
the difference between the average daily temperature of the 
outside air and 6o° F. in Britain or 65 F. in North America. 
In America it has been found that the actual consumption 
of fuel in central-heating installations is almost exactly pro- 
portional to the difference between 65 F. and the average 
outdoor temperature. For some purposes the standard tem- 
perature may be different ; in greenhouses a winter temperature 
of 45 F. may suffice, while various manufacturing processes 
may require temperatures up to 75 ° F. 

"Degree-days ," "degree-hours." — The basis of calculation used 
by engineers is the "degree-day." In countries using Fahrenheit 
degrees one degree-day is defined as a day on which the mean 
temperature is i°F. below the standard. Thus if the standard 
is 6o° F. and in one January the mean temperature is 40 F. 
and no day has a mean above 6o° F., the number of (60) 
degree-days in the month is (60— 40) X 31 or 620. Using the 
American standard the number of degree-days would be 
(65— 40) x 31 or 775. In a month in which some days had a 
mean temperature above the standard (i.e. degree-days =0) the 
value could be worked out from the mean temperatures of the 
individual days. This would be laborious even if the daily 
values were available, which is not always the case, and for 
such months J. R. Weeks (1942) gives the rule: For months in 
which not all days contain degree-days, subtract the average 
minimum temperature for the month from the standard (65 F.), 
multiply by the number of days in the month and divide by 



HEATING, AIR-CONDITIONING, LIGHTING, CLOTHING 235 

three. This rule is empirical, based on data for Maryland, and 
it is doubtful how far it is applicable in other parts of the world. 

In countries using Centigrade degrees a degree-day is defined 
as a day with a temperature i° G. below the standard. In 
Germany the latter was taken as 19 C.=66-2° F.; F. Bradtke 
(H. Reitschels, 1934) gives annual numbers of degree-days 
calculated on this basis for a number of places in Germany. 
To correct to ° F. they must be multiplied by 1 -8. However, in 
Germany it was customary not to begin the central heating 
season until the mean outside temperature fell to 12 C. (54 F.), 
and to shut it off when the average outside temperature rose 
again to this figure. 

The convention that the fuel consumption depends only on 
the daily mean temperature and not on the daily range is 
probably sufficiently accurate for large buildings, particularly 
for those occupied only during the day, but it is not likely to be 
true for small buildings such as dwelling-houses, outhouses, etc. 
For these a correct figure could be obtained only by studying 
a long series of hourly values of temperature. Apart from the 
difficulty that such data are available for only a very few places, 
the labour of the computations would be great. R. Grierson 
(1941) made the necessary calculations for a sample of days 
spread over five years at Kew Observatory, Richmond, Surrey, 
for each month for a series of standard temperatures rising by 
steps of 5 F. from 35°-70°, and he remarks that it would be 
impossible to deduce these figures for at least the lower standards 
from the monthly mean temperatures alone. Figures for other 
parts of the British Isles are to be taken from a map of 
"correction factors 55 to the Kew values. He also gives a series 
of correction factors for intermittently heated buildings, based 
on the number of hours which they are occupied each week. 

For the growth of plants the critical temperature is taken to 
be 42 F., and the Meteorological Office, London, compiles 
figures of the number of days during which the temperature is 
above and below this limit. These are termed "accumulated 
temperatures, 55 but are equivalent to "degree-hours' 5 divided 
by twenty-four. They are obtained by the following expressions : 

Maximum temperature below 42 F. : 42 minus mean temperature. 

Mean temperature below 42 F. but maximum above 42 F. : £(42— minimum). 

Mean temperature above 42 F. but minimum below 42 F. : £ (Maximum— 42) — 

£(42— minimum). 
Minimum temperature above 42 F. : no accumulated temperature below 42 F. 



236 CLIMATE IN EVERYDAY LIFE 

Accumulated temperatures above 42 F. are calculated in a 
similar way. Values for different maximum and minimum 
temperatures are given in Form 3300, issued by the Meteoro- 
logical Office, London. 

These expressions can be applied to find day-degrees below 
or above other standard temperatures, but the constants 
probably vary slightly. For a standard temperature of 6o° F. a 
better expression is : — 

Mean temperature below 6o° F. but maximum above 6o° F. : (60— minimum) /$. 
Mean temperature above 6o° F. but minimum below 6o° F. : 2 (Maximum— 60) /5— 
(60 — mimimum) /5 . 

These expressions can only be evaluated from daily values of 
maximum and minimum temperatures. It is desirable to find 
some simple expression from which the approximate number 
of degree-hours can be calculated from the climatic data 
readily available for most meteorological stations. 

If we have observations of shade temperature taken every 
hour over a period of years we can form a frequency distribution, 
which rises to a hump near the mean temperature and spreads 
out on either side. An example is shown in Fig. 29, in which 
the vertical scale represents the percentage frequency of occur- 
rence of temperatures within 0-5° F. on either side of the tem- 
perature in ° F. given by the horizontal scale. It refers to hourly 
readings of temperature in April at Kew Observatory. The 
ratio which the width of such a diagram bears to its height, if 
the horizontal and vertical scales are unchanged, depends on 
the variability of the temperature. A similar diagram for 
New York, for example, also for April, would be about one- 
third broader and only three-quarters of the height, because 
April temperature is more variable in New York than in 
London. 

For our purposes, however, a more convenient way to show 
the figures is by means of a curve known as an "ogive" or 
cumulative curve, which shows the total frequency below or 
above any temperature from the lowest to the highest (Fig. 30). 
From such a curve the number of "degree-hours" below any 
temperature can be obtained very simply by drawing a per- 
pendicular from the base line at the required temperature up 
to the curve, and measuring the area between this vertical 
and the curve (vertically shaded area in Fig. 30 with standard 
temperature of 6o°) . The area can be measured with a plani- 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 237 

meter or simply by counting the number of squares. If the 
curve is drawn on a vertical scale of i inch to 5 F. and a hori- 
zontal scale of 1 inch to 10 per cent., an area of one square inch 
is equal to 50 degree-hours per 100 hours or 360 degree-hours 




30 35 40 45 50 55 60 65 70 

Temperature °F 

Fig. 29. — Frequency distribution, hourly temperatures, Kew 
Observatory, April. 



per month of thirty days. The number of degree-hours above 
the standard temperature is given by the area above the curve 
to the right of the vertical line through that temperature (hori- 
zontally shaded area in Fig. 30) . 

Hourly readings of temperature are not available for many 
places, and even if they are the computation of frequencies is 



238 CLIMATE IN EVERYDAY LIFE 

laborious. An approximate cumulative curve can be con- 
structed without much difficulty if we have the following data, 
which are generally available for each month : — 

Mean temperature. 

Mean daily maximum and minimum temperatures. 
Mean monthly maximum and minimum temperatures. 
Highest and lowest recorded temperatures. 




30 



35 



T 

40 45 50 o 55 

Temperature *E 



Fig. 30. — Ogive or cumulative curve of temperature. 

The mean temperature should be an approximation to the 
mean of the twenty-four hours. This is given for many places, 
but the British and American practice is to take as the mean 
temperature the mean of the mean daily maximum and mean 
daily minimum. This is, on the average, about 1 ° F. too high, 
and 1 ° F. should be subtracted from the published mean if it is 
obtained in this way. 

In temperate regions we can, without great error, assume that 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 239 

half of the hourly temperatures lie below the mean temperature 
and half above. Hence we mark a cross on our diagram where 
the vertical representing the mean temperature intersects the 
horizontal representing 50 per cent. (Fig. 31). 



No daily range 
90- (24 xd&y degrees) 




30 40 SO 60 70 

Temperature °f: 

Fig. 3 1 . — Construction of cumulative curve. 



Roughly one-fifth of the hourly observations lie below the 
mean daily minimum, and one-fifth above the mean daily 
maximum. Hence we mark these figures on the horizontals 
representing the 20 per cent, and 80 per cent, lines. About one 



24O CLIMATE IN EVERYDAY LIFE 

in 200 of the hourly observations lie below the mean monthly 
minimum, and one in 200 above the mean monthly maximum. 
These values are accordingly marked on the horizontals for 
0-5 and 99-5 per cent. 

Finally, the lowest and highest recorded temperatures mark 
the extreme ends of the curve. Having plotted these points we 
draw a smooth curve through them and measure off from it the 
number of hour-degrees below or above any standard tem- 
perature required. For New York, April, we have (1930-39) : — 



Lowest recorded .... 


27 


Mean monthly minimum ... 


3i 


Mean daily minimum 


42 


Mean temperature .... 


49 


Mean daily maximum 


57 


Mean monthly maximum . 


79 


Highest recorded .... 


88 



In Fig. 32 the estimated curve constructed from these figures is 
shown by the thick line and the observed curve by the thin line. 
The agreement is excellent. 

All buildings smooth out the variations of external air tem- 
perature to some extent ; on the other hand, no building (except 
perhaps underground cellars) is so perfectly insulated as to have, 
if unheated, no diurnal variation of temperature. The daily 
range of temperature inside an unheated building is a fraction 
of that in the outside air, the magnitude depending on the size 
of the building, nature of its contents, thickness of walls, amount 
of ventilation, etc. 

Inside a house of moderate size the daily range appears to be 
from one-third to one-fifth of that in the outside air. Not many 
figures are known to me, but A.J. ter Linden (1938) gives some 
thermograph curves for houses in the Hague. I analysed one 
pair of these, comparing a room on the ground floor with the 
outside air, with the following results : — 



Mean max. ° F. 
Mean min. ° F. 
Range ° F. 



Outside air Inside room 

71-8 3 p.m. 68-0 7 p.m. 

61 -5 4 a.m. 65-8 8 a.m. 

10-3 22 



Both maximum and minimum inside the building lag about 
four hours behind those in the outside air. In a large factory 
in summer the daily range is about one-fifth or less of that in 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 24 1 

the air outside. Hence we can make an approximate correction 

of the curve in Fig. 31 to conditions inside a large building by 

shifting our 20 per cent, and 80 per cent, points C and E to new 

IOO, 1 , > 1 1 1 — ^==3P* iO 




60 



20 30 40 50 

Temperature °F 



70 80 



SO 



Fig. 32. — New York, cumulative hourly temperatures, computed 

and observed. 

positions C and E' nearer to the vertical through the mean 
temperature by an amount equal to four-tenths of the daily 
range. The whole of the curve from A to C is shifted to the 
right by the same amount and the whole curve from E to G 
is similarly shifted to the left. Between C and E a smooth curve 



242 CLIMATE IN EVERYDAY LIFE 

is drawn passing through D. Over most of its length this part 
of the curve will be coincident with the vertical through the 
mean temperature. The construction is shown in Fig. 31. 

The use of day-degrees implies that the daily range inside 
the building is zero. Hence to convert the curve of Fig. 31 to 
approximate day-degrees we mark the points C", D and E" 
all on the vertical through the mean temperature (broken line 
in Fig. 31). In practice the curve should be rounded off so 
that the central part is not quite vertical. The true curve 
certainly lies between CE and C"E". 

Maximum Load to be expected. — In addition to the number of 
day-degrees or hour-degrees of temperature below the standard, 
we require to know the lowest effective temperature to be ex- 
pected in order to assess the heating power which the installa- 
tion must be able to put out. The best figure to adopt for this 
purpose is the mean annual minimum temperature, as given 
in Appendix I. A map showing the distribution of this tem- 
perature over the world is given by C. E. P. Brooks and G. L. 
Thorman (1928). The temperature of the outside air falls below 
this for a short time every other year, but mostly at night, and 
the heat content of the building itself should be capable of 
carrying over these rare extremes. The heaviest demand on a 
heating system comes when a building which is only inter- 
mittently occupied, such as a hall, has to be warmed up for 
occupation during a period of exceptional cold. Since the 
building begins to lose heat as soon as its temperature rises 
above that of the air, the heating period should be as short 
as practicable, consistent with reasonable initial cost. For this 
reason the heating plant should be capable of slightly more 
than maintaining the required indoor temperature when the 
external air temperature is at the mean annual minimum. 

Other factors in heating and cooling. — The number of degree- 
days or degree-hours by which the temperature of the outside 
air falls below the required temperature is only one of the 
factors affecting the amount of heat required to warm a build- 
ing. We have to consider also : — 

Other sources of heat, especially solar radiation. 

Heat losses by conduction through walls, floor, windows and 

roof. 
Heat loss by warmed air escaping from the building. 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 243 

The warming of buildings by solar radiation was discussed in 
Chapter II (p. 57). The heat transmitted through windows is 
much greater than that through walls, and for that reason it is 
desirable to have as much window space as possible on the 
south side. Some types of glass transmit more radiation than 
others, and this point may be worth considering in cold, sunny 
climates. The possibility of conserving solar heat by day for 
release at night is being tested in the U.S.A. (M. Telkes, 1947). 
The conserving agent may be a large cellar water tank or a 
suitable chemical. 

Heat loss by conduction was also discussed in Chapter II 
(p. 75). The loss, like the gain, is greatest through windows, 
and as the loss increases with the wind speed (roughly according 
to the square root) it may be desirable, as far as is compatible 
with adequate daylight, to limit the area of windows on the side 
exposed to the most frequent and coldest winds, which in 
Britain is generally the north-west side. It should also be re- 
membered that on hill-slopes there is a flow of cold air down 
the slope on clear, calm nights. If a building stands directly in 
the path of such a flow it will be engulfed in cold air at night, 
but the airflow can often be held up or diverted by suitable 
wind-breaks (see p. 266). As most daylight enters through the 
upper parts of windows, heat can be conserved with minimum 
loss of daylight by keeping the windows high up in the walls on 
the side exposed to the coldest winds. 

The heat carried away by the warmed air leaving the build- 
ing is an important source of loss, which obviously can be mini- 
mised by cutting down the flow of air through the building. 
Most buildings in Britain are probably over-ventilated in the 
sense that an unnecessary amount of air passes through them 
without always circulating properly, so that the fresh air fails 
to reach some parts while other parts suffer from draughts. In 
the cold regions of the United States the modern practice is to 
cut down the flow of air through the building, but to maintain 
a good circulation within the building by fans and by suitable 
connecting channels. Walls, floors and ceilings are warmed 
and well insulated. Central heating is general; open fires are 
hardly known except as "decorations," and C. C. Handisyde 
(1947) suggests that this is because the abundant winter 
sunshine provides the necessary source of high-temperature 
radiation. 



244 CLIMATE IN EVERYDAY LIFE 

By far the greater part of the heating in Britain is done either 
directly or indirectly by the burning of coal, and much of this, 
especially in private houses and offices with open fires, is burnt 
in a very wasteful way. With the manifold increase in the price 
of coal in recent years this waste is no longer tolerable, and the 
general adoption of more economical methods of coal con- 
sumption is urgent. R. Fitzmaurice (1942) points out that the 
consumption of coal for heating per head in Britain was much 
greater than in Germany, where winters are colder, and yet a 
better standard of heating was maintained in German than in 
British buildings. The open grate has been discarded in almost 
every country of the world, except Britain, in favour of the 
closed stove. 

In countries with very cold winters the practice of heating 
a whole district from a single boiler is being developed, but 
Fitzmaurice doubts its practicability in Britain because of the 
great variability of our winter climate. The whole subject of 
heating small buildings is at present in a stage of transition and 
new developments are likely to emerge from the researches now 
being carried on. 



THE CONTROL OF TEMPERATURE AND HUMIDITY J AIR CONDITIONING 

The temperature of the air is not the only condition which 
affects the feeling of well-being and energy ; relative humidity 
and wind movement are also important. For this reason the 
concept of "effective temperature" has been introduced; this 
is defined as the temperature of motionless, saturated air 
which would induce the same feeling of heat or cold in a 
sedentary worker wearing ordinary indoor clothing as that 
given by the actual conditions of temperature, humidity 
and air movement. A number of charts of effective tem- 
perature at various air temperatures, relative humidities and 
air speeds have been published; a convenient one is that 
of the American Society of Heating and Ventilating Engineers 
(New York, 1944). In the U.S.A. the optimum effective 
temperature is regarded as about 67 ° F., but may range from 
65 F. in winter to 73 F. in summer. In Britain the optimum 
effective temperature is considerably lower, probably about 
6o°F. 

The air temperature, wet-bulb temperatures and relative 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 245 

humidities which give effective temperatures (E.T.) of 6o° and 
67 F. with air movement of 20 ft./min. are as follows: — - 



Effective temperature ° F. . 


60 


67 


Air Temperature, ° F. 
Wet-bulb, ° F. . 
Relative Humidity, % 


60 

60 

100 


61 

57 
76 


62 
54 
55 


63 
5i 
37 


64 
48 
19 


67 

67 

100 


68 

65 
86 


70 
61 
59 


72 
57 
35 


74 
54 
17 



Increased air movement lowers the effective temperature, even 
when the air is saturated. An air movement of 20 ft./min. is 
practically still air; an air movement of 100 ft./min. is appre- 
ciable, and this reduces an E.T. of 6o° to 57 or 58 F., and one 
of 67 to 65 or 66° F. For this reason, switching on an electric 
fan brings a feeling of coolness, and in moderately unfavourable 
conditions is a substitute for air-conditioning. 

Even at optimum E.T. very moist or very dry air is un- 
pleasant; the relative humidity should not be above 70 per cent, 
or below 30 per cent. The optimum temperature and humidity 
vary somewhat in different countries, depending to some 
extent on what people are used to. S. F. Markham (1942) finds 
that the ideal temperature in which men work hardest and most 
efficiently lies between 6o° and 76 F., and the ideal humidity 
between 40 and 70 per cent. Hence, in extreme climates the 
mere warming or cooling of the air is not sufficient; the humidity 
also needs to be adjusted, but not necessarily to a fixed figure. 

Fig. 33 shows in diagram form the comfort and danger zones 
in terms of air and wet-bulb temperatures. The operative part 
of the diagram is included between the two exterior sloping 
lines representing relative humidities of 100 and o per cent.; 
conditions outside these limits are physically impossible. The 
inner sloping lines show relative humidities of 70 and 30 per 
cent, and mark the limits of comfortable humidity. The two 
broken lines near the top of the diagram show the limiting con- 
ditions for heat-stroke calculated by D. Brunt (1947). Above 
the upper broken line heat-stroke is probable in a nude subject, 
resting in an air current of 17 ft./min., and above the lower 
broken line heat-stroke is likely to be brought on by moderate 
physical activity. For more active air movement (about 200 
ft./min.) the limit above which heat-stroke is probable is: in 
saturated air, dry- and wet-bulb about 95 F. ; relative humidity 
34 per cent., dry-bulb 122 F., wet-bulb 94 F. The curved 



246 



CLIMATE IN EVERYDAY LIFE 



IOQ 




60 70 BO SO IOO HO 

Air remper&rure°F 

Fig- 33- Comfort and danger zones. 



I20 I30 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 247 

line near the top of the diagram limits the temperatures above 
which heat-stroke is probable, according to R. G. Stone (1941). 
The horizontal line at wet-bulb 78 F. shows the limit for 
sustained white labour. 

The stippled area marks the comfort zone for acclimatised 
residents in hot tropical regions, and the shaded areas the 
accepted comfort zones in the U.S.A. and Britain. The middle 
lines in these shaded areas represent effective temperatures of 
73 , 67 and 6o° F. respectively. 

The process of maintaining a suitable temperature and 
humidity is known as "air conditioning." This also includes the 
cleansing or purifying of the air and sometimes introducing a 
suitable rate of air movement. In cold weather air conditioning 
implies warming and generally adding moisture to the air; in 
hot weather it means cooling and removing moisture. 

Air conditioning in cold weather. — In very cold weather the air, 
even if saturated, contains only a small amount of moisture, and 
when it is warmed up to a comfortable temperature it becomes 
uncomfortably dry. In such conditions, for maximum comfort 
water vapour must be added to the air. For example, if the 
outside air is at 20 F. and 80 per cent, relative humidity it 
contains only 1 -06 grains of water per cubic foot, and when 
warmed to 65 ° F. it will have a relative humidity of only 12 
per cent. The comfortable limits of relative humidity are 
between 30 and 70 per cent. ; if we adopt 50 per cent, as a 
standard, it is necessary to add about 2 J grains of water to every 
cubic foot or 5 -7 grams to every cubic metre of air brought 
into the building, after it has been raised to the requisite tem- 
perature. In meteorological tables weight of water vapour is 
generally given in grams per cubic metre ; this can be converted 
to grains per cubic foot by multiplying by 0-44. 

The calculation given above was based on the assumption 
that the building into which the warmed air is introduced does 
not itself contain any source of water vapour. That assumption 
is not usually correct. The occupants of a room themselves add 
water vapour to the air, mainly by breathing, and in addition 
many of the objects in a room can absorb moisture at high 
relative humidities and release it when the humidity falls. In 
practice it can be assumed that the relative humidity of the air 
in an occupied room in winter will be raised by 5 to 10 per cent, 
from these causes — the figure naturally varies according to the 



248 CLIMATE IN EVERYDAY LIFE 

number of cubic feet of space per person and the rate of ven- 
tilation. In rooms which are rather densely populated the 
occupants themselves are likely to maintain the relative humidity 
within the comfort zone. In less crowded rooms the occupants 
sometimes maintain the humidity by the crude but effective 
method of placing a saucer of water in front of or on the radiator. 
Air conditioning as distinguished from simple heating is there- 
fore relatively unimportant in cold weather. 

Air conditioning in hot weather. — Although the cooling of air is 
a comparatively recent development — it is much more difficult 
to cool air than to heat it — there already exists a large literature 
about its beneficial effects. H. C. Bazett (1948) remarks that in 
naval vessels in the tropics, blacked out and closed down for 
action, the air became io° F. or more hotter and also more 
humid than the outside air, and in submarines conditions were 
even worse, and led to a serious loss of efficiency. T. Bedford 
(1946) points out that the effect is increased by the crowding 
of modern warships with machinery, decreasing the available 
air space and requiring a larger crew to work them. After 
air conditioning was introduced into submarines the crews 
became the healthiest of the whole fleet. In very hot countries 
entering an air-conditioned room at a temperature much below 
that of the outside air is liable to give a chill. A sudden change 
in the effective temperature by 3 F. is tolerable, however, and 
in tropical factories the effect of cooling the air to 8o° F. and 
slightly dehumidifying was found to be very beneficial. The 
effect of air conditioning in a Chinese cotton factory was re- 
ferred to on p. 32. Outdoor workers gain in efficiency by 
sleeping in air-conditioned bedrooms, but this is only practic- 
able in large solid buildings such as blocks of flats ; the light 
open construction usual in many tropical buildings is not suitable 
for air conditioning. The tolerance of hot humid conditions 
varies widely even between men of the same race, and should be 
tested before a man undertakes regular work in such conditions. 

In Britain it is not often necessary to cool and dry the air 
even in summer — the chief need in large towns is to cleanse it. 
High temperatures are occasionally experienced, but the 
relative humidity on such occasions is nearly always below 60 
per cent., and more often than not below 45 per cent. F. H. 
Dight (1934) found that at Kew, London, in the thirty-four 
years 1900-33 the hourly temperature exceeded 85 F. at least 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 249 

once in nineteen of the years, the total duration being 232 hours 
spread over sixty-one days. The average relative humidity was 
41 per cent. The highest wet-bulb temperatures were 75-7° F. 
on 29th August 1930 (maximum air temperature 89-1° F., 
relative humidity 52 per cent.) and 75*5° F. on 20th July 1900 
(air temperature 89-4° F. relative humidity 51 per cent.). 
These two days also had the highest water contents in the air, 
17-1 and 1 6-9 gm./cu.m. respectively, and they may be taken 
as representing about the worst combination of high tem- 
perature and humidity likely to be met with in London. The 
highest air temperature of 93-9° F. on 9th August 191 1 was 
accompanied by a wet-bulb maximum of only 71*4°, the 
relative humidity being 30 per cent. Dight remarks that the 
most sultry conditions in Britain often come at the end of a 
warm spell when the air temperature is falling while the relative 
humidity is rising, so that the wet-bulb temperature changes 
little. These conditions usually end in a thunderstorm. S. F. 
Markham in a fascinating book (1942) points out that civilisa- 
tion developed first in those parts of the world which enjoyed 
an average climate nearest to the ideal, i.e. the sub-tropics. But 
in these regions there are many hours of the day, and even 
whole seasons, when the weather is too hot or too humid for 
full efficiency. Consequently as methods of heating improved 
and the winter temperature became less important, the centres 
of civilisation moved to higher latitudes, where the summers 
are less enervating. He attributes the rise of Greece to the dis- 
covery of the hypocaust system of heating, and of Rome to 
glazing and public baths. The fall of Rome was preceded and 
accompanied by the neglect of the housing system, and leader- 
ship passed back to Islam and the 70 F. isotherm. The suc- 
cessive discoveries of the chimney, coal, gas and electricity have 
made heating more and more easy and exact, and now the 
ideal climate is one which is just right in summer and can 
easily be controlled indoors in other seasons by heating — in fact 
the climate of England. 

The development of economical air conditioning, however, 
may reverse the process, and make the optimum climate for 
industrial purposes that which lies near the optimum for the 
greatest part of the year; Markham suggests a mean annual 
temperature of 6o° F. as possibly marking the greatest advances 
of civilisation in the future. 



25O CLIMATE IN EVERYDAY LIFE 



LIGHTING 



Artificial lighting has reached such a high level that there 
is a tendency to regard daylight lighting as unimportant. 
R. Fitzmaurice (1942) points out that while the requirements 
of different occupations in artificial lighting have been codified, 
and there is a strong body of professional skill to meet those 
requirements, few people bother about day-lighting, and he calls 
attention to a range of "daylight factor protractors" due to 
A. F. Dufton of the Building Research Section, for the rapid 
calculation of the daylight efficiency of any system of windows. 
The illumination at different distances from windows of various 
sizes and shapes with different skylines has been calculated in 
detail by T. Smith and Miss E. D. Brown (1944). They point 
out that the area of sufficient illumination at table height 
(2 feet 9 inches) from a window is an ellipse with its longest axis 
parallel to the window. For living rooms the window area 
should be sufficient to give one per cent, of the outside illumina- 
tion, but work requiring a good light should have at least 
2 per cent. They also emphasise the importance of high windows 
for three reasons. First, the area illumined increases with the 
window height; an unobstructed window 6 feet wide rising 
5 feet above table level gives an effective area for close work 
of 89 square feet, while one 2| feet high gives only 44 square 
feet. Secondly, obstructions such as buildings cut down the 
effective illumination much more with low than with high 
windows. Finally, daylight entering at a low angle (less than 
2 5 ) is more fatiguing than that entering at a high angle. 

The illumination is very much greater when the sun is 
shining than when it is obscured. The worst condition is a 
dense town fog, which often involves day-long lighting, but 
heavy cloud is almost as bad. Table 10 (p. 68) gives the 
average daily duration of sunshine in open situations for a 
selection of places. The duration is cut down by obstacles 
which raise the sky-line, especially between east and south- 
east, which obstruct the morning sun, and between south-west 
and west, which have a similar effect in the evening. 

Electric lighting is cheap, even nowadays, but daylight is 
free and is available for most of the working day. However, 
there are periods of dusk and night, and other periods of fog 
or heavy cloud, when artificial lighting is necessary, and the 



HEATING, AIR-CONDITIONING, LIGHTING, CLOTHING 25 1 

aim of the architect should be to make the transition from one 
to the other as smooth as possible, in the home, office and fac- 
tory. It has been found by experience that artificial light is 
switched on, on the average, when daylight has fallen to 12 
foot-candles, but is not switched off again until it has risen to 
100 foot-candles. This difference is probably due partly to the 
difference in colour, for the eye, adapted to the light of one set 
of wave-lengths, cannot instantaneously adapt itself to light of 
different wave-lengths. The transition could be made easier 
by the use of a system, such as fluorescent lighting, which 
simulates daylight. Fluorescent lighting also uses less current 
for the same candle-power. 

VARIATIONS OF THE LOAD ON ELECTRIC SYSTEMS CAUSED BY 

WEATHER 

The following notes are based mainly on a discussion at a 
joint meeting of the Institution of Electrical Engineers and the 
Royal Meteorological Society (London. Royal Meteorological 
Society. 1945). 

The use of electricity for heating depends mainly on tempera- 
ture, but is influenced also by wind speed, relative humidity, 
insolation and probably also by rainfall and intensity of day- 
light, since a dull, wet day may give the impression of being 
colder than a fine bright day, though the actual temperature is 
the same. The lighting load depends on the duration of day- 
light intensity below about 60 foot-candles. The loads for both 
heating and lighting vary regularly from winter to summer, and 
at different times of day, and this variation can be foreseen and 
provided for; but in addition there are rapid variations of load 
due to more or less sudden cold spells, the onset of strong winds 
and rain, fog and dense cloud. 

In the discussion E. B. Powell stated that the largest single 
factor was temperature, the load increasing slowly as air tem- 
perature fell from 8o° to 6o° F. and then more rapidly, but almost 
uniformly from 6o° downwards. With a temperature of 27 F. 
the heating load is estimated as three and a half times that at 
57 F. Curves shown by C. T. Melling were similar, and 
indicated that the increase of load per 5 fall of temperature was 
greater at temperatures below 35 F. than above that figure. 

The load due to lighting is small and almost constant so 



252 CLIMATE IN EVERYDAY LIFE 

long as daylight illumination remains above 60 foot-candles. 
Between 60 and 20 foot-candles it increases fairly rapidly 
(presumably because of the tendency of people to leave lights 
on after they are no longer necessary) ; from 20-0-5 foot-candles 
the load is almost exactly proportional to the logarithm of the 
difference between 100 and the actual daylight illumination. 
The load at 0-5 foot-candles is more than four times that at 20. 
A daylight illumination of 0-5 foot-candles in the open is prac- 
tically equivalent to darkness in a room, and at that level 
lighting is almost that of full night. 

Owing mainly to atmospheric pollution, dull, foggy days in 
London may have an illumination of only 14 foot-candles, so 
that lighting is necessary all day. The worst conditions occur 
during overhead fogs. In foggy weather there is almost always 
an "inversion" of temperature, the air near the ground being 
colder than that at a height of some hundreds of feet. This 
prevents the surface air from rising, and the smoke and fog 
particles tend to accumulate in the surface layer of air. But 
owing to the heat generated in London in winter the air in the 
streets and among the buildings is warmed up by two or three 
degrees (see p. 43), and this is enough to dissipate the fog at 
ground level. The smoke and fog particles all accumulate 
between this warmed layer and the level of the inversion, 
forming a dense pall through which hardly any daylight can 
penetrate, and full artificial lighting is required all day. 

In summer the greatest darkness is brought about by dense 
thunderclouds, which often come up suddenly and may be 
extensive. These may reduce the daylight intensity to as low 
as 4 foot-candles, which is considerably less than that shortly 
after sunset on an average day. This sudden load imposes a 
worse strain on generators than do falls of temperature, which 
generally come on more gradually. Moreover, forecasts of 
changes of temperature are issued by the Meteorological Office, 
but forecasts of daylight illumination are much more difficult 
and, apart from warnings of fog, have not, so far as I know, 
been attempted. 

The temperature and daylight factors can be combined into 
a formula of the form 

L=a—b log I—cT 

where L is the load on the installation, /is the daylight illumina- 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 253 

tion in foot-candles, T is the outdoor temperature (° F.), 0, b 
and c are constants, which can be determined from the records 
of any installation by statistical analysis. This relationship has 
been studied in detail by P. Schiller ( 1 944) . 



CLOTHING 

Clothing serves three utilitarian purposes, to keep the body 
warm in cold weather, to ward off rainfall, and to ward off 
excessive insolation. We can leave out of account here two other 
purposes, namely to serve as a receptacle for impedimenta and 
to adorn the body, as these are not connected with climate. 

Indoors excessive insolation and rainfall do not occur and the 
building ought to be kept at a comfortable temperature so that 
protection from cold should not arise. In occupations requiring 
considerable physical effort or which have to be carried on in a 
high temperature the problem is to keep the body cool and dry 
by getting rid of perspiration, and light loose-fitting garments 
are called for. In the open, however, protection from cold, rain 
and excessive insolation are all important. For a detailed 
catalogue of complete outfits of clothing, including footwear, 
to suit different types of extreme climate see D. H. K. Lee and 
H. Lemons (1949). 

Protection from cold. — The insulating power of a thickness of 
cloth is almost proportional to the thickness of the layer of 
"dead air" between the surface of the outer layer of cloth and 
the body. This includes not only the air spaces within the 
different layers of cloth, but also those between them. Closely 
fitting garments reduce the latter and are therefore unsuitable 
in cold climates. 

Wind lessens the insulating power of clothing in two ways — 
it penetrates the cloth and disturbs the layer of "dead air," and 
by pressing the clothes against the body it destroys the in- 
sulating layers between the different layers of cloth and between 
the latter and the body. The first loss of insulation can be 
lessened by wearing an outer covering of as low a permeability 
as is consistent with getting rid of perspiration. Completely 
wind-proof garments are not desirable because with no ven- 
tilation at all the inner garments become soaked and lose about 
half their insulating power. 

P. Larose (1947) examined a number of specimens of cloth 



254 CLIMATE IN EVERYDAY LIFE 

and found that permeability ranged from o to 193 cubic feet 
per square foot per minute under a pressure difference of J inch 
of water across the fabric. Since the weight of a layer of water 
\ inch deep and 1 square foot in area is 2-6 lbs., I take this as 
equivalent to a wind speed of about 30 m.p.h. directed normally 
against the cloth. He gives curves showing the thermal insula- 
tion in "clos" of cloths of different permeabilities from 8 cubic 
feet per square foot per minute upwards, at different wind 
speeds. One "clo" is the thermal resistance of clothing neces- 
sary to maintain in comfort a sitting-resting subject in a nor- 
mally ventilated room (air movement 20 feet per minute) at a 
temperature of 70 F. and a relative humidity of less than 50 
per cent. It is equivalent to a temperature difference across the 
cloth of 0-18° G. per gm. calorie per hour per square metre, or 
o-88° F. per B.T.U. per hour per square foot. 

In calm air the insulating power of all fabrics is about equal 
at 1-75 clo, and except for the most porous material this holds 
up to a wind speed of 6 m.p.h. When the speed rises above 
6 m.p.h. the curves begin to diverge, and are farthest apart 
with a wind of 20 m.p.h. At this speed the insulation of the 
least permeable fabrics (permeabilities 8 and 13 cubic feet) had 
decreased to about 1-2 clo, that of the most permeable (36 and 
193 cubic feet) to o-8 and o-6 clo respectively. Above 20 m.p.h. 
the insulation of the less permeable materials continues to 
decrease steadily, but that of the more permeable changes little. 
At 60 m.p.h. the curves would meet, so that in a very strong 
wind the nature of the material makes little difference. As, 
however, outdoor work would be difficult in winds much above 
20 m.p.h., and impossible in winds of 60 m.p.h., the latter con- 
clusion is of academic interest only. The author's conclusion 
is that for work in light and moderate winds it is of advantage 
to wear a light covering fabric which is nearly, but not quite, 
impervious to wind. 

The greatest need for protection against cold is in the Arctic 
countries in winter. Research into the best forms of protective 
clothing in Arctic Canada is described on p. 167. 

Protection against rain. — Coverings to protect the wearer against 
rain are of two types, waterproofs with a layer of rubber (mac- 
kintoshes) or oiled fabric (oilskins) which are completely im- 
permeable to air and water vapour as well as to liquid water; 
and water-repellant substances which permit air and water 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 255 

vapour to pass. The first type is proof against continuous heavy 
rain (unless the impermeable layer is cracked or worn into 
holes), but is equally proof against the escape of perspiration 
from within. The second type, which includes many "rain- 
coats," is showerproof, but not proof against continuous heavy 
rain; its advantage is that it permits perspiration to escape. 

Natural wool because of its greasiness and hairiness is a fairly 
good water-repellant and there are a number of patent materials 
which are better, but this book is hardly the place to discuss 
them. A "raincoat" in this sense is cloth coated with a hydro- 
phobic substance such as wax, but the air spaces in the cloth 
are not filled, so that it is permeable to air and water vapour 
to an extent depending on the size of the pores. It resists wet- 
ting by raindrops and rain does not form a film of water, but 
it permits the passage of water under pressure. For this reason 
heavy rain can penetrate a raincoat on nearly horizontal sur- 
faces such as shoulders, and driving rain can eventually pene- 
trate vertical surfaces. The best raincoat material can stand 
up to several hours of rain and even then pass water slowly. 

An investigation into the properties of water-repellant sub- 
stances was carried out by the National Bureau of Standards, 
Washington (J. W. Rowen and D. Gaglierdi, 1947). The 
authors found that the most important factor is the angle of 
contact which the surface of a drop resting on the fabric makes 
with the surface of the fabric. If this angle exceeds 90 , so that 
the shape of the drop is larger than a hemisphere, the contact 
between drop surface and fabric is on the outer side of the 
latter; the inner surface remains dry and the fabric is a good 
water-repellant. If the angle of contact is less than 90 the shape 
of the drop is a flat segment of a sphere, contact between the 
surface of the drop and the fabric is on the inner surface of the 
latter, and the water-repellant properties are poor. 

A number of tests are available for assessing the efficiency of 
water-repellant fabrics, some of which measure the resistance 
to penetration of water under pressure, and others resistance to 
the impact of drops, but none of these is completely satisfactory. 
The "drop-penetration" test appears to be the best because it 
approximates most nearly to the natural conditions. 

The value of a water-repellant cloth depends not only on its 
resistance when new, but also on the extent to which it de- 
creases after wetting and subsequent drying, and also after 



256 CLIMATE IN EVERYDAY LIFE 

cleaning. The resistance of some fabrics was found to be nearly 
halved after five wettings and dryings. 

Protection against insolation. — The sun's rays may be divided 
into actinic (producing chemical changes), light rays, and heat 
rays. All the sun's rays are actinic to some extent, but the most 
active are the short "ultra-violet." It is these rays which cause 
inflammation of the skin (erythema) and sunburn. When the 
skin becomes pigmented, as in sun-tan after the inflammation 
has died down, the pigment forms a protection against further 
inflammation. 

At one time the actixncj^avs^were believed to be the cause of 
many of the troubTeTluchas heat-stroke, sunstroke, conjunc- 
tivitis and nervous diseases which affect white settlers in the 
tropics, but they are now regarded as Jess dangerous thanj he 
lig ht and heat rays. Nevertheless, they probably contribute to 
the effect of the latter and some precautions against them are 
desirable. Sir Aldo Castellani (1938) quotes Woodruff as 
advising that in the tropics the outer clothing should be white, 
grey or yellow, which reflect a large proportion of the light 
rays, and underclothing should be black or yellow, which he 
thinks stop the ultra-violet rays (this is not certain). The 
danger from actinic rays is greatest in dry, tropical countries 
such as Egypt; in moist heat the water vapour in the air stops 
a large proportion of the ultra-violet, and in fact in such con- 
ditions many whites do not tan, but their skin assumes a peculiar 
whitish tinge. This does not seem to have any ill effects. 

The intense light of tropical regions is responsible for various 
eye troubles, especially tropical photophobia with headaches 
and neuralgia, glare conjunctivitis, night blindness, etc. Pro- 
tection of the eyes by adequate headgear and by dark glasses 
or Crookes' non-actinic glass with side-pieces is desirable. In 
towns much of the glare is due to reflection from white build- 
ings, and this can be minimised by colouring the walls green or 
brown instead of white. 

The infra-red or heat rays in solar radiation are now con- 
sidered to be the main cause of sunstroke and sun-exhaustion. 
They are absorbed by the body and especially by the scalp, 
causing failure of the heat regulation. Castellani advises all 
Europeans in the tropics to wear broad-brimmed pith helmets 
or topees covered with white or khaki cloth and lined internally 
with red, yellow or black. There must be free ventilation and 



HEATING, AIR CONDITIONING, LIGHTING, CLOTHING 257 

free circulation of the air inside, as otherwise the head would 
become very hot. The effectiveness of the helmet is increased 
if a sheet of aluminium foil is moulded over the top, covered 
with a white cloth merely for appearance. He described experi- 
ments which showed that the temperature under such hats 
exposed to the sun was several degrees lower than under similar 
helmets without aluminium foil. White is, of course, by far the 
best outside colour, air inside ordinary white helmets being 
some 20° F. cooler than under similar hel 
For the effect of colour in cloth see p. 178. 



CHAPTER XII 
ALTERING THE WEATHER 

IN this chapter we deal briefly with some attempts to change 
the actual weather in the open. The scale of processes in the 
atmosphere is so great that hitherto man's purposefully 
directed efforts have had only small and local results; the unin- 
tentional effects of human activities such as extensive deforesta- 
tion on the climate have been much greater, but unfortunately 
they have usually tended towards a deterioration of climate. 

The various attempts may be classified under the following 
heads : — 

(i) Rain-making and fog dispersal. 

(2) Prevention of hail. 

(3) Protection from lightning. 

(4) Frost prevention. 

(5) "Wind-breaks." 



FOG DISPERSAL 

The possibility of inducing rain to fall by artificial means is 
at present a matter of controversy. It is generally agreed that 
rain cannot fall unless a cloud is present ; the question is whether 
in certain circumstances an existing cloud can be made to give 
rain when, if left to itself, it would not. 

Cloud particles are formed when the amount of water in the 
air exceeds that which can be held as water vapour. This 
happens when moist air is ascending and cooling by expansion. 
The surplus water vapour condenses on nuclei in the form of 
water droplets, which are visible as cloud. In absolutely clean 
air the water vapour would not condense readily, but would 
remain as vapour in a state of super-saturation. In nature, 
however, sufficient nuclei of condensation are almost always 
present. J. R. Ashworth (1929) considered that there was some 
evidence that rainfall in the manufacturing town of Rochdale 
was least on Sundays, when the mills are closed, but in the 
discussion of his paper the evidence was not regarded as strong. 

So long as the temperature of the cloud layer is above 32 F. 

258 



ALTERING THE WEATHER 259 

the cloud will consist entirely of water drops of small and 
generally uniform size, which have the same electric charge 
and do not coalesce readily to form raindrops. If the tempera- 
ture falls below 32 F. the water drops do not immediately 
freeze unless some ice-particles already exist in the air. Instead 
they continue to exist as "super-cooled" water, and can main- 
tain this state at times down to a temperature of — io°F. 
(— 23 C.) and even lower, but they cannot retain the liquid 
form at a temperature of — 3i°F. (— 35 C), and a super- 
cooled water drop at any temperature below 32 ° F. freezes as 
soon as it touches a particle of ice. 

In air which is just saturated and contains only water drops, 
the latter will be in equilibrium, i.e. they will neither grow by 
condensation nor decrease by evaporation. If some of the water 
drops are turned to ice by contact with ice-crystals this equili- 
brium is disturbed. Air which is in equilibrium with water 
drops is supersaturated for ice, so that water vapour condenses 
and freezes on the ice-particles. The loss of vapour from the 
air is made good by evaporation from the remaining water 
drops. The ice particles grow at the expense of the water drops 
until they are large enough to fall out of the cloud. If the air 
below is cold enough for them to remain frozen they fall as 
snow; if they melt on the way down they turn into rain. 

It follows that save in exceptional circumstances a cloud 
cannot give appreciable rain unless (a) it extends into the 
freezing level in the atmosphere, and (b) ice-crystals are present 
on which raindrops can form. Clouds which do not extend 
into the freezing level rarely give more than fine rain or drizzle. 
The exceptions occur chiefly in tropical regions where the 
freezing level is very high, but the air contains so much moisture 
that large raindrops can form without the intervention of ice. 

The modern theory of rain-making is that the place of the 
natural ice-crystals can be taken by "seeding" the cloud with 
small pellets of some substance which is either very cold or very 
hygroscopic. In practice the substance usually employed is 
"dry ice" (solid C0 2 ) scattered from an aircraft flying just 
above the cloud. The pellets of "dry ice" falling through the 
cloud cool a thin tube of air to below the temperature of — 3 1 ° F. 
at which water drops necessarily freeze without the intervention 
of ice. Each pellet therefore leaves behind it a trail of ice- 
particles, which are spread by turbulence through neighbouring 



260 CLIMATE IN EVERYDAY LIFE 

parts of the cloud. T. Bergeron (1949) points out that a nice 
adjustment of conditions is required for an appreciable amount 
of rain to be produced by this process. In the first place, 
the cloud must be dense and thick to contain enough water to 
supply the rain. This can be the case only if there is a con- 
siderable depth of cloud at heights below the freezing level. 
Secondly, the cloud must extend above the freezing level or 
any ice-particles produced would quickly melt. Thirdly, there 
should be sufficient updraft of air below and through the cloud 
to support the growing ice-particles until they reach a good 
size and also to maintain the supply of moist air necessary 
for prolonged rain. These three conditions are met with only 
in large cumulo-nimbus or thunder clouds. But such clouds 
generally grow sufficiently high to reach the level at which ice 
crystals exist naturally, which is usually at a temperature of 
15-20 F., so that they would produce rain without artificial 
"seeding." In most experiments in which seeding by solid CO a 
has produced rain there is no means of knowing whether or not 
rain would have fallen in any case. Some meteorologists hold 
the view that seeding makes no difference. However, in an 
experiment in Australia (E. B. Kraus and P. Squires, 1947), 
out of many hundreds of cumulus clouds present, two were 
seeded and these two were the only ones to produce rain. The 
question is still open, but at present it seems unlikely that rain- 
making will have any appreciable economic importance. 

In some cases seeding by solid C0 2 has the opposite effect of 
dissipating thin stratus cloud, and Bergeron remarks that this may 
be useful for aviation and at coastal holiday centres. Fog dis- 
persal, i.e. the local clearance of fog from airfields, which was 
much discussed during and shortly after the war, depends on a 
different principle. By the expenditure of a sufficiently great 
amount of fuel it is possible to keep the temperature of the air 
over a small patch of country above the dew-point and hence 
free of water fog. The cost is so great that its use is only justified 
in exceptional circumstances. Fog clearance is not likely to have 
any industrial or commercial applications. 



PREVENTION OF HAIL 

Hail is formed when a pellet of ice alternately falls into the 
level of liquid super-cooled water drops in the cloud and is 



ATMOSPHERIC POLLUTION 26 1 

carried up again into the region of ice-crystals. In the former 
it acquires a layer of hard, clear ice; in the latter it grows by 
the accretion of crystalline ice which contains a good deal of air 
and is more or less opaque. Large hailstones often consist of 
several pairs of these concentric layers, which can be seen when 
the hailstone is cut open, and the structure has been described 
as resembling that of an onion. Each pair of layers represents an 
upward and downward journey across the boundary between 
the level of super-cooled water and that of ice-crystals. Large 
hailstones require a strong upward current of air to support them 
(see p. 210) and hail is formed only in violently ascending 
currents such as those which accompany thunderstorms and 
tornadoes. 

The weapon most usually employed against hail is the "hail 
cannon," the theory apparently being that the shock wave pro- 
duced by firing the cannon vertically upwards penetrates the 
layer of super-cooled drops and shakes them into freezing. In 
the frozen state they are no longer available to add to the size 
of hailstones. A less plausible argument is that the shooting may 
break up the rising current which bears the hailstones aloft. 
According to Col. Ruby (1938), in the wine-producing regions 
of southern Europe the layer of super-cooled water drops often 
lies between heights of 5,000 and 10,000 feet and the shock wave 
from a hail cannon reaches a height of about 6,000 feet, so that 
the hail shooting may have some effect. In very severe storms 
the super-cooled layer extends above 10,000 feet and there is 
no shock effect. 

Hail shooting was first tried in the seventeenth century in the 
French wine district of Beaujolais, but its use did not become 
general until near the end of the nineteenth century. In the 
heyday of hail shooting 10,000-20,000 shots were sometimes 
fired in a single storm. The practice is now decreasing, but is 
still in use in some places. The decline may be attributed to the 
fact that while the cost is certain the benefit is uncertain. 

Rockets bursting at 3,000-4,000 feet have sometimes been 
used instead of cannon, but they are more costly and their effect 
has not been systematically investigated. Col. Ruby states that 
they are especially effective against hail-bearing tornadoes. 
The latest development along these lines is the bombing of the 
super-cooled layers by aircraft. Rockets were released from an 
aircraft into the super-cooled layer during a hailstorm in July 



262 CLIMATE IN EVERYDAY LIFE 

1937 and the hail ceased, but more experiments will be necessary 
before any real effect can be regarded as proven. 

The second method of combating hailstorms is the "electric 
niagara." This consists of a group of high masts erected on hill- 
tops and well earthed, the theory being that these conductors 
would dissipate the difference of potential between earth and 
cloud which accompanies a thunderstorm. A two-year investi- 
gation, however, failed to show that electric niagaras had any 
effect at all on the precipitation of hail. 

The third method in use is the ionisation of the air. Positive 
ions from radio-active material are directed upwards by a 
positively charged ball, the idea being that they will enter the 
ascending current of a hailstorm and in some way interfere 
with the development of hail. Some of these instruments have 
been installed, but it is not yet known whether they have any 
effect. Similar ideas such as radio-active rockets and bombs 
have been suggested, but not tried out. 

The difficulty in all these operations is that the phenomena 
are on so great a scale that the cost of changing the course of 
nature may exceed the damage to be feared if nature is left 
alone, so that they are not economically sound. Moreover, it is 
always difficult to prove that the result obtained would not 
have followed in any case without human intervention. 



PROTECTION FROM LIGHTNING 

Damage and destruction by lightning are due mainly to the 
sudden intense heat caused by the passage of the discharge 
through poorly conducting materials. The thermal expansion 
ruptures brick, concrete and similar materials and the heat sets 
fire to anything inflammable; fires are also caused by the dis- 
charge jumping gaps between two conductors. Dislocation of 
delicate electrical apparatus is a different effect, caused by the 
sudden surge of current. 

The principle of the protection of structures against lightning 
is to provide the discharge with a safe and easy passage to earth. 
Since metals are the best conductors, lightning rods are made 
of metal, and the highest point should be connected to earth 
by at least two widely separated conductors, which should be 
as straight as possible. Sharp bends are especially to be avoided, 
as lightning is always ready to take short cuts. For the same 



ALTERING THE WEATHER 263 

reason any large metal objects in the building, especially those 
within 6 feet of the lightning conductor, should be connected 
with the latter, or earthed, or both, to prevent side-flashes. 

The highest point of a lightning conductor is the apex of 
a "cone of protection," which is an imaginary cone with a 
vertical axis, extending to the ground like a tent. All points 
within the cone are supposed to be protected by the lightning 
conductor, but there is some doubt about the ratio which the 
radius of the base bears to the height. The Washington Bureau 
of Standards "Code for Protection against Lightning" (1945) 
states that the radius of the base of the cone is from two to four 
times its height. R. H. Golde (1946) quotes a number of cases 
in which buildings were struck at points within the 2:1 cone, 
but he states that these were weak flashes which did little direct 
damage. The ratio of the radius of the cone of protection to its 
height increases with the strength of the flash. 

The protective ratio also decreases with the height of the 
conductor. This is because the precise point to be struck is not 
determined until the leader stroke is about 50 feet from the 
ground, i.e. below the tops of high towers, and at this level it 
may turn aside and strike sideways. The lowest protective 
ratios observed, i-i:i, were in experiments in Russia with 
earthed balloons when the latter were at a height of 900 metres 
(2,950 feet). Golde also states that the protective ratio of an 
earthed horizontal wire is less than 2 : 1 and probably less than 
1-5:1. 

As the leader stroke approaches the ground there is a rapid 
flow of electricity through the earth to the point immediately 
beneath it, and this earth-current may be sufficiently strong to 
cause damage. Hence it is desirable to earth the conductor in 
at least two points on either side of the building to be protected. 
The best earth is a metal conductor, such as a water-pipe which 
can carry the earth-current safely from a considerable distance. 

Although weak lightning flashes may sometimes strike build- 
ings protected by outside conductors, they rarely penetrate the 
interior. Interior damage is generally caused by induction, due 
to the electrical installation being too near the conductor, and 
is usually confined to burning out a few lamps and blowing 
fuses. In unprotected buildings, on the other hand, the electric 
wiring often offers the path of least resistance to earth. The 
greatest danger is from aerial conductors entering the building; 



264 CLIMATE IN EVERYDAY LIFE 

Ch. Forel (1939) considers that currents in such wires, due 
either to direct striking or to induction, caused most of the 
damage reported in protected buildings in Switzerland. He 
recommends that where practicable aerial wires near the 
building should be replaced by subterranean cables of large 
capacity. 

PROTECTION AGAINST FROST 

Frost, while especially feared by farmers and fruit growers, 
also interferes with many other activities. Frosts are of two 
kinds, general frosts caused by wide, deep currents of air below 
freezing point sweeping over the whole countryside, and local 
frosts due to radiation from the ground on calm, clear nights. 
The first type of frost, which comes with north-east or east 
winds in Britain and north or north-west winds in North 
America, is often termed a "black frost" because the air is dry 
and there is little or no formation of hoar frost. Such frosts may 
continue for several days and nights, and very little can be 
done to counteract them. 

The second type, radiation frost, is due to the loss of heat 
from the ground to the air and sky on clear nights with calm 
air or light winds. The ground chills the air above it and the 
layer of cold air flows down the slopes and accumulates in the 
hollows and narrow valleys. The air is usually cooled below 
the temperature at which it is saturated, and moisture is de- 
posited on the ground and vegetation, but since these are below 
freezing point the deposit takes the form of hoar frost; conse- 
quently a radiation frost is often termed a "white frost." 

Various methods of combating radiation frost have been 
tried. These are described by R. Bush (1946), and in very great 
detail by O. W. Kessler and W. Kaempfert (1940). They may 
be classified as : — 

(1) Preventing loss of heat by radiation. 

(2) Stirring the air to bring down heat from above. 

(3) Diverting the flow of cold air. 

(4) Heating the air by burning fuel or spraying with warm 

water. 

Preventing loss of heat by radiation. — Since the cause of frost is 
the radiation of heat from the ground, crops, etc., the fall of 
temperature can be minimised by intercepting the heat radiated 



ALTERING THE WEATHER 265 

and returning it to the ground. This can be achieved by 
covering the crop to be protected. The covering may take the 
form of a layer of loose straw, cones or ridges of straw, etc., 
or one of the various types of glass "cloches." These may keep 
the temperature 3-5 F. above that of freely exposed objects. 

On a somewhat larger scale protection can be given by the 
erection of a temporary "roof" resting on poles a few feet above 
the ground. The construction can be quite light, but there 
must be flaps which can be dropped on the up-wind side to 
prevent cold air from drifting underneath the cover. This form 
of protection is sometimes used on a small scale to prevent the 
freezing of cement in process of setting. 

It has long been known that serious radiation frost does not 
occur on cloudy nights. From this comes the idea of forming an 
artificial "cloud" by burning some form of smoky fuel or by 
one of the various chemical smokes available — the nature of the 
particles does not make much difference. A really dense cloud 
can maintain a temperature 4-7 ° F. above that in neighbouring 
areas not covered by the cloud. At one time this method was 
in great favour, but it is now being abandoned. A concentration 
of smoke dense enough to be effective is also dense enough to be 
a nuisance, and even harmful; in the U.S.A. the amount of 
smoke which may be emitted by a heater is now limited by law. 
Further, on sloping ground the flow of cold air from higher 
levels can cut under the smoke cloud, which then loses most of 
its protective value. 

Since water vapour absorbs radiation and re-radiates it, the 
suggestion has been made that the addition of water vapour to 
the air, for example by burning damp straw, would cut down 
the loss of heat by radiation. The effect of water vapour, how- 
ever, does not depend entirely or even mainly on the amount 
in the lowest layer of air, but on the whole mass of water vapour 
in the atmosphere above the locality. There are various other 
difficulties, and it is very doubtful if the gain of temperature by 
this means can exceed i° F. 

The statement is sometimes made that a smoke cloud or 
similar protective cover raises the temperature by so many 
degrees. This is incorrect; all it can do is to decrease the rate 
of fall. 

Stirring the air. — In a radiation frost there is an "inversion" 
of temperature. The air is coldest near the ground and becomes 



266 CLIMATE IN EVERYDAY LIFE 

warmer at higher levels ; at a height of 50 feet above a level 
plain it may be several degrees higher than at ground level. If 
this warm air could be brought down to ground level the result 
would be an appreciable rise of temperature. This method of 
frost prevention is theoretically possible on a level plain or 
gentle slope, but not in a frost hollow where the layer of freezing 
air is much deeper. There is another advantage: in still air 
radiating surfaces fall below the air temperature, but in moving 
air this difference is minimised. Experiments are being tried, 
but so far the results are not conclusive, though there has been 
a slight rise of temperature in the immediate neighbourhood 
of the ventilator. According to the Daily Telegraph for 16th 
May 1949 investigations with horizontal fans were being 
carried on near Silsoe, Beds and Royston, Herts, for prevention 
of frost in orchards. 

Diverting the flow of cold air. — Any obstacle placed across the 
flow of cold air on a sloping hill-side results in a local accumu- 
lation on the up-slope side, and makes a small "'frost-hollow." 
A high hedge, wall or row of trees on the slope below a garden 
or building therefore increases the risk of frost, unless a gap is 
cut in the barrier to allow the cold air to drain away; a similar 
obstacle on the slope above holds back the cold air and de- 
creases the risk of frost. This effect is most noticeable where the 
hill slope is interrupted by a terrace or slight hollow. A hedge 
planted obliquely down the hill can deflect the flow of cold air 
to one side. The obstacle must extend down very nearly to 
ground level, for the air is coldest near the ground and can 
easily penetrate a belt of trees with trunks bare to a height 
of 10 feet or so. Where for any reason walls or hedges are 
impracticable, or while hedges are growing, temporary screens 
may be hung on posts in strategic positions when there is a 
risk of frost. This method of protection does not decrease the 
loss of heat; it merely re-distributes the cold air to places where 
it can do less damage. 

The effect of gaps in an obstacle is quite remarkable. It has 
been noticed, for example, that where a railway embankment 
follows the contours of a slope, but is pierced by an opening for 
a road, on the up-slope side the height to which trees are frosted 
is lower on either side of the opening than farther away, while 
on the down-slope side there is an area of frost opposite the 
opening, but none on either side. 



ALTERING THE WEATHER 267 

In planning an orchard, garden, group of storehouses, etc., in 
which it is necessary to minimise the risk of frost, much can be 
done by careful siting and arrangement after study of the local 
air drainage. On the other hand, as Bush points out, it should 
be possible to make use of cold air flows to keep goods cool in 
summer by building storehouses across the night flow of cold 
air, preferably below ground, opening them by night and 
closing them by day. 

Heating the air. — The most effective means of protecting a 
garden or orchard against frost is by actually warming the air. 
Any form of heating can be used, the most practicable being 
oil, coal or peat. The choice depends partly on the relative cost 
of the fuel on the spot and partly on the labour available, since 
oil burners can be arranged to burn all night, while coal stoves 
require more frequent attention. Efficient heaters can main- 
tain a temperature 7-10 F. above that of unprotected areas. 
A report on the efficiency of different types of heater has been 
prepared by R. Gallay and P. Darbre (1948). 

The arrangement of the heaters is a matter of considerable 
importance. A single source of great heat has little effect 
because the heated air breaks right through the inversion and is 
lost in the upper air. The effect to be aimed at is a local cir- 
culation of the air within the inversion layer, so that the heat 
is not lost. The air over the heater rises and is replaced by air 
drawn in at ground level on all sides, but at a moderate height 
the heated air is drawn sideways and down to replace the sur- 
face air. This effect is best obtained by a number of small 
heaters. For normal situations Bush recommends fifty heating 
pots to the acre, which is about one per hundred square yards, 
the number to be increased in bad frost hollows. Kessler and 
Kaempfert recommend 1 to 60 square yards. Burners pro- 
viding 10,000 kcal./hour (about 40,000 B.T.U. per hour) 
spaced at the rate of 1 to 60 square yards can raise the 
general air temperature by 6-8° F. Actually, burners should 
not be equally spaced; in still weather they should be closer 
on the edges than in the middle, and if there is any breeze they 
should be closer on the windward side. 

When the frost is caused by the flow of cold air down slopes 
a "barrage" of hot air has been tried. This takes the form of 
either a row of heaters across the slope or a gently inclined iron 
tunnel of inverted U -shape, in which air heated initially to a 



268 CLIMATE IN EVERYDAY LIFE 

very high temperature at the lower end flows up the slope of 
the tunnel, warming the upper surface of the iron. 

Oil burners are very popular in the U.S.A., where oil fuel is 
plentiful and cheap. Even with these advantages, however, 
heating is expensive and uncertain. Not infrequently the cost 
of the fuel is greater than the value of the crop saved. 

Where facilities are available, either from permanent stand- 
pipes or from water-carts, spraying with water is said to be 
effective. Water acts in three ways: first, the temperature and 
especially the heat content of the water is higher than that of 
the air ; secondly, if the water freezes, the release of latent heat 
checks the fall of temperature at just under 32 F., which may 
not be low enough to cause damage; thirdly, the conductivity 
of the soil is increased by wetting so that more heat escapes from 
the sub-soil to warm the air above. Of these the second effect 
is the most important. Kessler and Kaempfert state that a 
suitable amount of water to release is 5 litres per square metre 
per hour, equivalent to about 5 J quarts per square yard. Too 
much water swamps the ground. Experiments in Germany 
resulted in the temperature being held at 3i°F., but the 
method has not yet been sufficiently tried out to say whether 
it is really efficient. 

Forecasting frosts. — In view of the cost in fuel or labour or both 
of anti-frost measures, it is essential that they should only be 
taken when really necessary. This involves forecasting the 
occurrence of frost a few hours ahead. Warnings of the risk of 
frost are included in the official weather forecasts, but in con- 
ditions favouring radiation frost it is the local topography which 
determines whether any particular spot will be in a danger area. 
Various methods have been devised for making local forecasts. 

The simplest method is based only on temperature. If, fairly 
early in the night, the thermometer falls below 35 F., frost is 
probable before morning. There are a number of devices 
available which ring a warning bell when the temperature falls 
to 35 F. (or any other level pre-selected as dangerous). These 
may consist of a mercury thermometer with two contacts sealed 
into the glass, apparatus depending on the different rates of 
contraction of two rods of different metal, or other similar 
arrangements. If some of these are placed in the most dan- 
gerous spots and wired to the house, early warning is given. 

Even on a cloudless night the rate of cooling depends very 



ALTERING THE WEATHER 269 

much on the humidity of the air. Owing to the latent beat 
liberated when dew or hoar frost is formed, the night minimum 
rarely falls much below the dew-point — probably 4 F. is a fair 
allowance only exceeded in exceptional conditions. Conse- 
quently an idea of the probable night minimum can be obtained 
by reading a psychrometer consisting of a dry-bulb and wet- 
bulb thermometer, finding the dew-point from tables which are 
often supplied with the instrument, and subtracting 4 F. (If 
the dew-point is below 32 ° F. hoar frost will be formed, and it 
should then be termed the "frost-point".) 

Temperature and humidity do not exhaust the list of factors 
affecting the risk of frost. Cloud is obviously important ; on a 
cloudy night frost is unlikely if the temperature at sunset is not 
below 38 F. Other things being equal, frost is less probable if 
the soil is wet than if it is dry. Wind speed is important; a strong 
wind breaks up the surface layer of cold air and prevents it from 
drifting into the valleys. Wind direction has to be considered ; 
even a moderate breeze blowing against the slope of the ground 
minimises the risk of frost. Finally, the length of the night is an 
obvious factor; starting with the same temperature at sunset, a 
lower minimum is to be expected on a winter night than on a 
night in spring, the difference in western Europe being roughly 
i° F. for every ij hours difference in the length of the night 
from sunset to sunrise. A discussion of the effect of humidity, 
state of soil and length of night was given by L. Dufour (n.d.) ; 
who calculated the following figures for the fall of temperature 
over dry soil at Brussels between sunset and sunrise : — 



Relative humidity at sunset % 


100 


80 


60 


40 


Probable fall of temperature : ° F. 
Night of 16 hours 
Night of 10 hours 


22 
18 


23 
19 


25 
21 


29 
24 



These figures for relative humidities of 100 and 80 per cent, at 
least seem to me to be improbably high, but local topography 
dominates the variation of temperature at night to such an 
extent that general tables are not of much use for special cases. 
Experience on the spot, aided by keeping records of self- 
registering maximum and minimum thermometers and dry- 
and wet-bulb thermometers read at a fixed hour each evening 
is likely to be the best guide. 



270 climate in everyday life 

"wind-breaks" 

The use of rows of trees as a protection against wind in open 
country is a very old practice ; if properly managed they are 
also ornamental and a useful source of timber and fuel. In 
recent years a number of investigations have been carried out 
which show that this method of protection is effective. The 
most comprehensive were by W. Nageli (1943, 1946) in various 
parts of Switzerland where wind-breaks are erected as a pro- 
tection against the strong winds which sweep up the valleys by 
day. Investigations in Russia were carried out by B. A. Bodroff 
(1935). For a discussion of the use of wind-breaks in a sub- 
tropical country see E. J. Kelly-Edwards (1945). These and 
other investigations all show similar results. Practical hints on 
growing wind-breaks are given by A. A. Pardy (1946). 

The best form of wind-break is a belt of mixed trees five to 
ten yards wide, which contains at least three rows of trees and 
is moderately dense. Spruce, which was formerly used ex- 
tensively in Europe, is too dense while young, and as it ages it 
tends to develop bare trunks beneath the crowns, allowing the 
wind to sweep through near the ground. Deciduous trees alone 
are too permeable in winter when they lose their leaves. The 
best compromise is a mixture of the two, which should be 
managed in such a way as to form a barrier of equal perme- 
ability from the ground upwards. The height should be as 
uniform as possible, except at the ends or at any unavoidable 
gaps, where the level should be tapered off to avoid the formation 
of turbulent eddies. Wind-breaks should, of course, be planted 
at right-angles to the locally prevailing winds. 

Four zones may be distinguished in the variation of wind 
speed on either side of a wind-break. In Fig. 34 the horizontal 
scale represents horizontal distances up-wind and down-wind 
from the barrier, expressed as multiples of the height of the 
latter. The vertical scale represents wind speed as a percentage 
of the speed in the open plain away from all obstacles. Up-wind 
from the barrier (zone A) the wind speed begins to decrease 
at a distance equal to about six times the height of the barrier. 
Immediately behind the latter (zone B) the wind speed falls 
to a very low figure of 15-40 per cent, of the free wind, the 
lowest speed being at an average distance of three or four times 
the height. The denser the barrier, the nearer the minimum is 



ALTERING THE WEATHER 



271 



to it, the smaller the wind speed, and the steeper the following 
rise. The third zone, C, extending from about six to about 
twelve times the height, is that of rapid recovery of wind speed 
to a value of 75-80 per cent, of that in the open. Here the wind 
is often turbulent, the turbulence being greater the steeper the 
rise, i.e. the denser the barrier. The turbulence brings down- 
drafts of air which sometimes flatten crops, and for this reason 
too dense a wind-break is a disadvantage. Finally, in zone D 
the wind gradually returns to its undisturbed condition; the 




Fig. 



Ratio, distance Prom break: heighT of break 

34. — Wind speed up- and down-wind from a wind-break. 



latter is usually reached at a distance of twenty-four to thirty 
times the height. 

These ratios appear to be almost constant, irrespective of the 
height of the wind-break and the strength of the wind. For 
example, a belt of trees 50 feet high protects five times as large 
an area as a hedge 10 feet high, and if at any spot a wind of 
20 m.p.h. is reduced to 10 m.p.h., a wind of 10 m.p.h. will be 
reduced to 5 m.p.h. (Bodroff, however, thinks that the sheltering 
effect increases with the strength of the wind.) If the wind is 
very turbulent the reduction factor is smaller. This probably 
accounts for the fact that in a succession of similar wind-breaks 
the second and subsequent ones are sometimes less effective 
than the first. According to Nageli, the protection extends 
to and slightly above the level of the top of the wind-break, 
decreasing little with height above the ground, but I should 
expect the recovery to open conditions to be rather more rapid 
in the upper levels. 



272 CLIMATE IN EVERYDAY LIFE 

If a system of wind-breaks is being planned, it follows that 
they should be planted at distances apart of about twenty-five 
times their expected height. Thus, if they are intended to grow 
to 50 feet they should be not more than a quarter of a mile 
apart. Allowing for each belt to monopolise a strip some 
10 yards wide, this means using 2-2 J per cent, of the ground 
area for wind-breaks. The protection from wind and decreased 
evaporation makes the sacrifice of this amount of ground 
profitable for agriculture in windy districts. 

The effect of a wind-break is to slow up slightly the whole 
mass of air passing through and over it, but this is compensated 
to some extent by increased wind speed at the ends of the 
barrier and probably also at some height above it. At the ends 
and behind any gaps of appreciable width the average wind 
speed may be 20 per cent, higher than in the open, but this 
local increase does not persist far down- wind. In designing 
wind-breaks care must be taken not to place them in such 
positions that they hold up flows of cold air and so form frost- 
pockets in places where frost can cause damage (see p. 79). 

CONCLUSION 

As previously stated, the whole problem of controlling the 
weather is bound up with the supply of energy and the location 
of points of attack. The mechanical energy of a large mass 
of rapidly moving air, such as constitutes a cyclone or baro- 
metric depression, is so great that the cost of the power necessary 
to exert any appreciable effect on it would be prohibitive. 
R. Corless (1930) computed that in the severe storm of 12th 
January 1930 the energy crossing a vertical section of the 
atmosphere of one square mile (i.e. crossing a line one mile long 
up to a height of one mile) near the Channel Islands was about 
100,000,000 kilowatts per second. If this storm could have been 
produced by expending electricity the cost at one halfpenny per 
unit would have been £750,000,000 per hour. The thermal 
capacity of a large mass of air, and the electrical energy of a 
thunderstorm, are similarly too large for human efforts to have 
much effect on them. The only possibilities of modifying 
weather at a relatively small cost lie either in attacking the 
chain of cause and effect at its weakest link, as in rain-making 
and hail-shooting, or in gently guiding the natural phenomena 



ALTERING THE WEATHER 273 

into channels where they can do the least harm, as by the use of 
lightning conductors. 

If by the development of atomic power the supply of available 
energy at small cost is increased many hundredfold, the case 
will be different, and large scale control of the weather may 
become possible. One such possibility, breaking up the Arctic- 
ice cap with atomic bombs in order to decrease the storminess 
of the North Atlantic regions, has already been canvassed in the 
Press. The results of such action, however, would be very com- 
plicated and might as easily be disastrous as beneficial. If 
civilisation endures, mastery of weather will probably come, 
but not without much thought, trial and tribulation. 



APPENDICES 



APPENDIX I 
CLIMATIC TABLES 

THE following pages give climatic data for a represen- 
tative collection of places. They are mostly based on 
the official publications of the countries concerned, and 
have been collected over many years ; some of them appeared 
in my earlier book Climate. They should, of course, have 
included information about the hours of observation and the 
period covered, but this would have doubled the space required 
without being of much help to the non-meteorologist. As 
temperatures are given to the nearest whole degree and rainfall 
to the nearest tenth of an inch, which are sufficient approxi- 
mations for the purpose of the book, the omission is of 
less consequence than if the figures had been set out more 
minutely. 

Most of the headings are self-explanatory, but a word is 
required about "Mean Annual Maximum (or Minimum) 
Temperature." The Mean Annual Maximum is obtained by 
finding the highest daily maximum temperature in each of 
a number of years and taking the average. It represents the 
highest temperature to be expected in an average year, and 
will be exceeded in about one year in two. The Mean Annual 
Minimum is obtained in a similar way from the lowest daily 
minimum temperatures in each year. These are among the 
most valuable data of applied climatology. 

Humidity is represented by the Relative Humidity, as 
nearly as possible the mean for the twenty-four hours. From 
this and the mean air temperature any other measure of 
humidity can be found, such as water-content or dew point. 
These can be calculated roughly from the figures in Appendix 
II, or more exactly from detailed tables such as the "Hygro- 
metric Tables" of the Meteorological Office, London (1940). 

I had intended to include frequencies of days of snow and 
fog, but the available data proved to be too scanty and not 
comparable for different stations. References to snow and fog 
have therefore been included only in the "thumb-nail sketches" 
or "Climatic character." These, on the right of the climatic 

277 



278 APPENDICES 

tables, are something of an experiment. They have been 
obtained as follows: — 

Starting with temperature as the basis, we can first of all 
classify the different places into : — 

Hot. — All months have a mean temperature exceeding 65 F. 
Warm. — At least one month below 65 , no month below 40 ; 

annual range exceeds 30 F. 
Warm equable. — At least one month below 65 °, no month 

below 40 ; annual range less than 30 F. 
Warm extreme. — At least one month above 65 , at least one month 

below 40 F. 
Very extreme. — Difference between warmest and coldest month 

exceeds 50 F. 
Cool equable. — All months below 65 , no month below 27 F. 
Cool extreme. — At least one month above 50 , at least one month 

below 2 7 F. 
Cold. — All months below 50 , at least one month below 27 F. 

The effectiveness of rain depends on the temperature at 
which it falls. An inch of rain dries off very quickly in a hot 
summer, but in a cold winter it has a lasting effect on soil, 
roads, etc. We may roughly classify climate according to the 
way in which the annual rainfall R in inches is related to the 
mean annual temperature T in ° F., and the season at which 
rain falls, as follows : — 



Rainfall at all seasons 
Rainfall mainly in winter 
Rainfall mainly in summer 



Dry 



R below 
T/2-12 
R below 
T/2-15 
R below 
T/2-10 



Moderately 
Rainy 



T/2-12 
to T-25 

T/2-15 

to T-30 

T/2-10 

to T-20 



Rainy 



T-25 to 

2T-50 

T-30 to 

2T-60 

T-20 to 

2T-40 



Very 
Rainy 



Above 
2T-50 
Above 
2T-60 
Above 
2T-40 



Where the mean annual temperature is less than 40 F., 
evaporation is generally so small that the amount of rainfall 
ceases to have much meaning, especially as in winter it falls 
entirely as snow. For convenience in such cases, we describe 
a rainfall of less than 10 inches as a dry climate and one of 
10 to 20 inches a year as snowy. An annual rainfall of less 



APPENDICES 279 

than 5 inches is described as very dry and one of less than 
2 inches as nearly rainless. 

If the mean annual relative humidity exceeds 80 per cent., 
the climate may be described as humid ; if it is less than 50 
per cent, as arid. Similarly, if the mean annual number of 
days of snow, thunderstorms or fog exceeds 50 the climate 
may be described as snowy, thundery or foggy, and if the 
number exceeds 75 these conditions are extreme. 

London (Richmond) has a mean temperature of 63 ° F. in 
July and 41 ° in January, and is cool equable. The rainfall is 
well distributed and amounts to 24 inches a year with a mean 
annual temperature of 50 F., so that the climate is moderately 
rainy. At Washington, D.C., the mean temperature ranges 
from 77 in July to 33 F. in January (warm extreme), annual 
mean 55 , rainfall 42 inches well distributed (rainy). At 
Rangoon (Burma) the lowest monthly mean temperature is 
77 (hot); annual mean 81 °, rainfall 99 inches mainly in 
summer (rainy, dry winter). 



280 



APPENDICES 





J 


X 


3 


Q 


H 


£ 


o 


W 


H 


Pu, 


2 


Ph 


§ 


< 


J 



u 



d"c3 



H O 



Ih Ih t-i 



as 



asa 



I'll 



.a .a .a 



4> 4) 4) 

333 

a a a 
crcrcr 
<u v 4> 



4) 4) o> 

333 
a a 3 

o* cr cr 1 
a) 4> a) 



4) 4) 0) 

333 

3 3 3 

crcrcr 

4) 4) 4) 



3 3 3 

4> 4> 4) 



3 3 3 
4> 4) 4) 



o o o 
o o o 

uuu 



o o o 
o o o 



o o o 
o o o 
OOO 



o o o 
cScScS 



o o o 



£.9 

.as 



iK 



o a 
6£ 



OMOO C» ""> M VO TJ- U"> fN VJOO N *t -*0O 



! fell 



« ftft 






ft <3 ft 



9 as ft s 



II 



*a 



fO fs 0\ nooi ** 7* -0 
i N « co ro N oo co • 



[J O O-M O->Jbj0 ^J+j'+j* obco w 
•- < 4) 4> O 0>O3 OOO 4)34) <>? 

QQO QO< OOO Q<!Q Q 



o O 00 

« w a 



V s v y 

f^ Tf ^1- 
N N M 



OMO 00 



CO 00 ON^O 



7*- 

ob 


U"> CO 


a 


>' 

O 


• — >' — > 


hi 
3 
< 


m 


<£>\0 


00 


M 


fOVO 
CO N 


M 



|i.HMM Q\ 0\ , T)-OCI 



l« « I 



VO M <^ 00 



0\ 00 0\ 00 o\ 



O ^O 00 ■<*• 

tx^o c* o 
OWO co oo 



££w £££ ££w £££ £ £££ 

"o "o "o\ *co v o\"fo <b V)V> lb Vi oo v h vb V> v h 

lOMM MH 0» Tf M NC0HN 

OOO OOO OOO OOO O OOO 

H CO H IOO N MHW N (OH \Q U")O0 O 

fcjzjjz; fcisfc £££ ^'^Z fc fcfcfc 

b>oo Vs *b\oo "o\ oo "h *io "o Vi"b\ *m Vi'tOn 

nn WW <*■ <*> m m M CO ■* <N 

CtMM O w CO O m« NmO 00 +M (O 

minm mmio iomm uo>n*© v-> ifimm 



W WW W W 
So \j-Vi "n "o 



£ ££ fe £ 



3 3 4) 

o o-a 

111 

WW2 



•S-o 

2 O 3 

aa§ 

ili 

ftH>< 



^3-S 



a, 
.8 . 

A 

o 

o 



ilia 



I - I 
I 1 « 



IS 



CQ CQ 



APPENDICES 28l 



i i 



6 * 
IP & 



£ £ bo°& a ,;&& 






U ft Mil ill fli 11 in Hi 

I lb jf a § js |§ 23 § ^3,0 §3 jg« « s£& . gs .55 



a£ 



1^ I fifi ^^s ag-s 1 ^ as* Hi HI 111 -Baa 



^ s-~ 



,,„§£ "3 "3^o "3 00 §§§ §c33£ §§ oS 8 C^S -3-3 g'g , 



OlMOO CO M H IO NO Ol^ CO O Ol fO ON H t<l H t)-CO ■«*■ ON O N OlO VO CO 

OhO M m h n mm hum O O h HO n h h hhh hhO OOO h h 

.ob'-a a b b d b" b Sf d d £\-V 3 e* £> ,0! ,d d .d ^ d -d ,d j^ J?? Sfi? 3 xi 

hSh >-, S S h SS -«; 1— >i — , h-ji-j W h-iH) pinpn r-, W<>-> WW>-i !-,<»-, WW 



op hco vp vo co vo coop T*" vf> ?"> P^T* P^P vp n h O •«!• •»*• HO* oou->p M Y"> 

N ff)« N CO CO VO COM CO in CO ^- in N CO^t- CO CO CO CO CO CO u-> COVO N N V COCO 

3^i^ d too &*> .b o ti ti +J d +j .M-Md-M >• > >> bb bi ti b >> >-§> > > d oi bi 

S3 3 3 p 4> 33 O 4) O OOHO IO 33s3 333 33O OOD 33 



eo pen 00 op h n vo h mpio inp H ioh rj- y-> ^j. focnop pvop h fs n n co h 

6 +Oi O CO co vo <xh OOob N h Ji N vo co co ^ 4j-ob rh •ifu-iio in 0\ Ov OOO 

NNH N N N IO IN N CO ■<)■ N NCO n vN (X !ON m N N N CO VN CO HHH N M 



H OMO 

NVO VO 




00 


co 
IS 


CO 


CO Ov 




10 vo 


CO 


VO fx 

VO *N 


co <y\ 
Nvo 


vo 


OsvO P» 
>OKN 


N O H 

vo tN.^s 


00 


-+o 

u-)VO 


NN 


CO **• H 
OOOOO 


00 
00 


00 

00 


O 


CO 

00 


00 0\ 

00 00 


VO CO H 

00 00 00 


00 co 
vo VO 


VO 

00 


•*h 
00 CO 


CO u-> 

CO CO 


CO 
CO 


ir> 0\ 0\ 
CO CO CO 


OOCCN 


CO-*H 
tN (X tN 


00 Oi 
00 00 


P* CO CO 


H 


1 


T 


^o 


H CO 
1 1 


O U"> H 


CO »N 


VO 

H 


0\ 1 


M vo 


H 


Ov O co 


H CO CO 

1 " 


Cv 


H 00 

VN CO 


00 H 


H OlM 
O\00 CJ« 


CO 

00 


CO 
00 


in. 

00 


CO 


00 00 


VO N rh 


N O 

0>O\ 


0» 


CO 1 
0^ 1 


O H 

OlOv 


O 


H VOOO 

vTvCO CO 


VO H CO 

00 Ov Oi 


O 

c 


O Ov 
O O^ 


VOOO 

0000 




O 


>o 


co 
VO 


IT) 


N O 


CONOO 
VO VO VO 


NN 


VO 


vO * 
vo N 


CO ■* 

vo vo 


vo 


IT) CO IO 

VO VO vo 


vovo N 




00 


Ovoo 
tsts 


VO vo 


>o 

M CO CO 


CO 


N 


O 


CO 
CO 


covo 


H ■* IO 
"* * CO 


CO •* 


CO 


CO * 


CO CO 




CO 


*rOO 

CO CO CO 


Ov vN in 
CN CO 10 


00 




00 VO 
CO CO 






O 


O 


CO 


O «"> 

•»<■ CO 


*NH 

10 vr> u-> 


r^vo 


in 


O H 

10 vo 


OvtN 


00 


O\«N00 

"* * * 


VO K * 
<+ -*vo 


COMIO 

VO vO vo 


Ooo 

IO Tj- 



NOM vOiOTt- voco *N N 00VOH OvvO O HOO U Ov O 

O00N CO VO ■* N 08 in voco-* -^-HVO 1010 100OCO rnio N co^ov voovco hw 

00 OVVO Tj-Tf--»J-00 CON mvo hhh rt- «h * coco txco co 

VO covO M C* « ^J" 



w w p4 ^ w'ni ^^pq WW W WW WW W WWW WW^ WWW WW 

"co 00 tO ".- tvoo V iN 00 *io V)0O v Ov vo ■<*■ ThOO 10 O 00 co ■>*• O H coco Ov ■>$- h 

o^ o"* o* 5 ^o™ o^o"" o^c^ 1 o M 0^0^ o o H o oVo" o r °o M o' N o^o^o o^o"" 

0» ■* vo ? n •* 10 Oh * 10NN (NOO VOcoN oOOvN mOvvo cONio cou-> 

HNVN^NN HH HM NNN 



*h "co in H vo CO N 



OOi* CO CO Ov 

m co •<*■ h •+ tj- 

000 00 

■* ov in co co 00 

■* Tj- rt- +* * 



5? V .. . . . . . . H • • . . . ..*j.. 

lfil &' JiaS.li 1.. !ol c B i|.a» A* .aSSift 



2111 ill! i illil ill! I§1 llllll Hi » 



282 



APPENDICES 



•i I 



ft -9 



'c3 ti 



gft 3 

erg cr 



&1 






•p 


O 


.3.3 

c3 CD 




g.s8 

>2> 


•d-d 








111 


laa 


x 2 


0> i> <U 


s * * 

w <U 4) 



■a & 



■II 



a 2 aa-3 -a 



a 



* « a 

cycrx 



S 3 3 « rt 



£££ £ 



§* 


en o 


> fl 


>;>; 


If 






<u <u 


aa 








« * 


IP 4) 




ss 



■3 2 



oSc 



H I 



CO 

w 

PQ 

5 

o 

1 

U 



<p\ fN N N co^p H op 9 «nrOO\ 9 9 °p 

M M N 6 OH H 66 CO H M HMO 



9 9 +* ■* o* *• °p 9^ 9 

H MOO O 6m mmn 






fl.o 
W 






3 a 



S3 S^."? 



•5 c 



^ g ^ M^feS W^S 



^3 



"3-33 



^S 



g d.o 



II 



+J -M O -^ -M 



bjC 6 w Mrv 

3 o> o 3 57 



-J d > 

0)0 



a) o 



N COCO tS 

co >o o m 



00 M 
M N 



St 3 



, <N VO 



I III 



G\ 0^ C^* 0^ O O* 



i oo 00 ChOO 



3 « 
4) H> 



i* I 



w £ ww w 



WW 
"b\ v o 



www WWW £ W* W*££ W WW www 



M rj- <r> C« CO 

oo o oo ooo 

VO CO N tJ-oo tNVOOO 



£ fc £££ ££ £ £Z £££ zzz 



£ £Z£ £ £2 £££ 



* 5 3 



it 



111 111 



l4 #jj4fijj]jtit4 mm i 



K> 



Sis « 
fill II 



APPENDICES 283 




°n 1 


8 


-a 


8 § 

a a 


c 

a 

u 


0) 

a 
g 


X . * 


a 


* 




§ 


a 


a ° a 




a 


£ £ 


a 


£ 



h B 

•3 1 2 -g 



-55 ? 'I Sa " a 
a"o ."> 000 o o ^i?>S 




H fx TJ- ON OlH 0\ VO CO N H O 9O M IT) H CO N Th fHO ?N£nH 7*" T*" H 

H O OO OhO O M h O O O O o *■ OOO 00 N N OOh mwO 

§ <! feS *=>&,&, hS h W ►=>£, W ^i$WO OT <! APhS^Hh $ W<! Q^Q QQfe 



CO fx fNCO lOOH IN Op O ^~v — op w 1— — 1 — O O v ~v — ^ — 1 — co O 00 vO t-, in CO H •**• 

Tt-OOOJOOoONoootij- T*" CO °? *? N ON r > ^.?*" » lilfl OiOm NNN 

O OH hvo^mi-imhhm 

^3§_3&3§^§o ^b ft » o r -§g S S -d « ^ ^33 3 33 



CO 

00 


vp 
on 


VO ON 


CO rfip 

on in 


On 

b\ 


vp 


«n 

CO 


00 




0\ 


CO 


9 
b 


N 


vp 
M 
CO 


CO 

00 


p 

On 
Ol 




00 

b 

ON 


vp On On 
VO -+00 


MNM 
00 4h H 
CO -a- 01 


O 
vo 




00 00 

10 vo 


MHO 

r-. txvo 


m 


CO 

VO 


m 

VO 


CO 

VO 


ft 


O 


VO 


m 


00 


IX 


■n 

00 


O 

ON 


ON 


00 


1 H OO 

1 00 00 


co ■* m 

0O0ON 


On 


00 

00 


<>s 

0000 


00 VO 00 
00 00 00 




to 

CO 


00 






00 


N 

N 


VO 





00 


m 
vo 


ts 


vO 


On 




00 


1 00 tx 


NNU1 


* 


o\ 
c*» 

1 


T T 


H M 

1 


VO 


■* 


00 


00 
VO 


1 


vO 
if 


On 
in 


m 


N 


00 


Ol 


ON 


vO 


in 

VO 


O «+ CO 

-<j- co M- 


■ vO co 


On 


5 


m in 

000 


H (V W 

OnoO On 


m 
Ov 


ON 


CO 

ON 


CO 

O 


1 


VO 
O 


in 

O 


O 
O 


CO 

On 


O 


m 

ON 


O 


On 


0» 

ON 


On O On 


I On co 
1 OO 


■* 
N 




VO 


(x co 

Nvo 


NO CO 


VO 
(x 


fO 


- 


00 
00 




ON 


00 


ON 


00 


in 

00 


OO 


5 


On 


00 


•* 00 c» 
00 00 00 


N h O 


•* 


00 


OOO 


CO Tj-fv 
co h c» 


CO 


Jn 


CO 


VO 


CO 


VO 


rx 


O 


O 
CO 


o\ 




co 




tx 


On 


Onoo O 
in -<j-vo 


vo 00 m 

CO CO M 


00 

10 


CO 


Old 


00 O O 


in 
m 


to 


0» 

>n 


CO 

00 


00 

m 


00 


00 


On 
in 


00 


O 

00 


01 

00 


00 


00 


O 

00 


txvo N 


O ONTJ- 



vo 0) invo 



1 N 00 ■>*• 



w 


W 


w'w 


WWW 


W 


W 


w 


w 


w 


W 


W 


W 


w 


W 


w w 


w 


w 


WWW* 


w w w 


00 


*■« 


MVO 


NO* 
















V H 






V> H 





V N 


*roO 


tvvO O 


































in 
















































00 O 






























COVO Tj- 




CJ 


Tt- 


•>*■ 00 


•* CO CO 


■* 


N 


M 




VO 


in 


CO 


vo 




ON 


On On 








M H H 



"ci "Vo "h vo vo vo v o~n Vo 00 n on V. v o v oi v o "o "b ON ^O vo *b %(■ W Vioo "coVj v o 

i»i win **N ■* * + <♦ co hoo Hin m ■* •* (O m hcoh hh 

h -sj- vo On 00 invo H-st-in N -*vOH Oo hhvO M vO coCTiCJNho 



r ' "i 



MO In JI1 --kn J 3 ili!lliii!,i!!l J 

b 



id** 



^ M B|l I aiwil lllill ill ills |3al ^^^ 



284 



APPENDICES 




jf s'-a 



$i £ i I 

! fl | I | 

a «> a a Is 
£ > £ £ £ 



In 



> & 



VO , VO CO ȣ 
O J 10 VO J* 



M P 



06 6 6 






<u On 2 0, <uOy OrtO 



6 6 6 6 

o S o 3 



lit 



JUS 



<u o 



< 



£*' 



bb >> ^1 



_b> u bi 

3 O rt 3 



OS 



<U VU +J 

r a> 3su » 



& a 



CO 

PQ 

3 






■si 



0O» Oi 00 



<N f-» t>« CO VO 



s s 



0» h o> o\ 



? I 



f^ cotx 00 tr> 



3 
II 





CO 


VO 


VO 







* 


O 
O 



0^ H tX M VO 



vovo a m 



W WW WWW WW W 



WW WW 



www w w 



CO w 



£ ££ £££ ££ % 



Z £ fc £ 55 £ £ £££ £ £ 



0) ONOO 
MM* 
OOO 



* I 



■ a • • 



LSI 

PI 



"S s 



5 >* 

mm 



■a J 

wS 



ON 43 



■1 



m h 9 m m 



3 
« a) rt m 



Ik 

II 



APPENDICES 285 



8 5 A 

.3 >, u >> ? 



53 



■a > . . 'a "a I 

M . o £ ^ £ . M . £ 

t3 C3 bo £ £ T3 73* • e 

O f> be O O o O r"""-" 

a>£ s s a a£2 



. as & & as ^ ag & 2 -g & sgaaaaaaaaa. 

■g i '3 ii ii as ; a 2>as>d > a a a g ana a B a BaBu 

u x u x x x Rs g ~ « » •• - B _r - 6 56 xxx xxx xxxjf 

■g » -g o> « © a a a oo o "So a § -g a fcj rt ^^ « • » «^° 

K>K>>>££WKKKK£cS ffi £ £ £ >>> > > > >>> 









CO 




VO 


vo 


Oi m 


IT) 


CM O 


00 


o 


o\ 


cm 






N 


00 OO 


Ch"<j-tH 







r- 




CI 


lO 


■* 


lO CM 


rO 


"<*■ Ch 




ch 


j% 




u-> vo 


0\ rJ-CO 


O 


t-. o 


lOCOOl 






w 




1-1 






M CM 
















H 




M 





co. • y-> in cm vo cm o r h r ?"'? , Y" > Y' ) 7" >l r' c / , P w ?' > 

6 ° ° 6Co6 coo 6 _? b _? bo bob b °_ b bob 



4 $ i 4 HH H i 44 t is Sfg- i- . - ' 



Ch M M- 



I* 



rj hb d "£, +J > d ►>, j J3 ><dd i? ob >> bb bb >> h >> bb 

D ' <" u sou 3 © qjoO o> ~h 3 «3 <S « m a P a 3 3 3 ,j3 « 3 



.CO rh Ch CO CO 9 Ch CO Ch 9 ^n in N ff, COOOOi O 03 N H 

I h V in wn ui a* * co >b b mco coo Af 



fNtv 



o 

OO 


OO 


in 

OO 




O 


1 


t^N 


o 


oo t^ 


CO 


00 t^ 


00 
VO 


ir> 


VO 


CO 


* 


■+ 


in co cm 
vo NN 


tv 


co 


O 


CM 00 Ch 
rj- moo 


vo 

00 


lO 


oo 


VT) 


N 
f^ 


1 


tN 


it 


^N0o 

t^OO 


vO 


00 tx 


o 


a 


00 


VO 




vO 


W CM CM 

00 00 00 


O 

00 


Ch 
vO 


CM 

00 


NON 
ts C^vo 


Ch 

VO 


N 


VO 

vo 


CO 

T 


1 


1 


^ 


Ch 


<o o 


1-^ 


VO O 
VO ts 


o 


! 


VO 


0\ 


VO 


CO 

c* 


•* m ■<*• 
1 1 1 


O 

o 
! 


vO 
CO 

1 


1 


m cm m 

1 OO H 
' 1 1 


Ch 


<3\ 


Oh 


oo 


Ci 


1 


CO 

o 


00 

0\ 


oo rh 
0>fj> 


CO 

o 


OiO> 


u-) 


1 


CO 


co 
O 


CO 


o 


HOlOl 

Osco CO 


Ch 


tv 


1 


N ION 

Ooo 00 




it 


00 


VO 


tStN 


00 




O M 
00 00 


00 


Ov 0\ 


CM 

00 


VO 






tx 


O 

00 


Ixvo ■»!- 
vO vo vo 


CO 


CO 


vO 


O O m 

oo vo vo 


t>. 


in 


Oh 


c^ 


1 


VO 






1O00 


Ol 




tv 


M 
CO 


O 
oo 


VO 


CO 


r^ 


CM O>vo 
1 1 1 


T 


T 


" 


gvo 
m 1 


On 


ft 


00 


in 


oo 

CO 


- 


c^ 


CO 

U3 


o o 

00 oo 


CO 

oo 


00 0\ 


o 


00 


O 

00 


vO 


CO 


VO 


co 0\ O 
CO CM CO 


>n 


<N 


CO 


vO CO O 

m ^- 



vOMinMvOCh CMm INOt^iNCMM oo oo 00 OOMtxcMinoooo Choo O 

CMCOCh-^-CMvO lOCO Tj-CM tSCMM CO ■* •>*■ HT*-cO^CO00MTtvOCMCM 

CMCMHinCh Tj- mcm m m ChcMmcMm Chmco^ 



wwwwww WW wwwwww w w w wwwwwwwwwww 

*b 00 *Ch "co "o vb *Ch "co "ch"ch "o Wl oo "cm %*• ^N 00 *Ch "KVi "ch "o 'K 00 00 W 

"I «1 ■+ N Ttci inn tncMco cmcm co co tJ- ThMwinMcowcMin 

vOvo cocovots r)-in OmOooo >n m oo m CMtxcocM-^-ocoo ChcoM 

O CM O CM cmoO COCO CMCM O ChO CO Ch CM CO cOCMOOChO t>«Tt-vovOcoco 

oS £££%'&£££;£&£ jESoS £ £ a^ £ z> £ £££ z ^ £ ^izz 

m on vo oo V) v cm oo "K "inoo V> V>vb "V So "b V. oo vo *d 'Kvo "m *m "o "d "Vo V. 

m cm m -<J-tj-tj- •>^--<r o+ + ciin m •* w m 10 « cmcmm ^j-cMincMco 

vOfvHMincO cmm rhChcocOOco 0\ O °m ChOO coco CM •«*■ Ch vO mNM 

co •^Tj-rj-cocoM m co cm hi ro cocoinminvomin ■fO ■* 



fe 



< 



9 -a . s o 



ft 

♦j rt*; 



iiiiitytoli iiiiiliiiiii liitiiiffli hi Hi 

§m t5»-,^c««S sop's W ^5Swlm|S w ?ffl* j|« g < k < S«J««A S O ^ h>> 



2 86 



APPENDICES 



20 8 B*8 



2 .a .s 



3-P 3 

trxcr 
met) 



i"l I 

u X 2 X! g 

*, <u a v a 



0) d V 

asa 


4) 4) 

an 


a 




















«i!x 


x x 


x 








>u>" 


>> 


> 



.a .a 



v ©.a 

a a 2 

g 8 J 



O Q « 

uow 



"II 



as 



o> 



hi 

X 2 B 

a 31 



rt O oJ 



coco tx o >£> 



too m 



tOOtO vo 00 to 



co com 6 6 






4) 4) C3 

3 3 y 

3 3 5* 



3^ 









3 > o 

a o 4) 




4) >s 

-a 

4) R 



v^> M 



t^t^ NVD 



I I 



f 1 



•n co | co m 
I I ' I I 



-« M Ol O 



pL| M O co »r* 



S2 



I I 



m l^vo co ' 



0\ ■* M O co 
MNN co. 



■* VO O 



£££ £ £ £ £££ ££ £ £££ ££ £ £ £ £ £ ££ 



V> v o Vi *io Vi V- V)"b\V» ovo So V-V"b Vco Vj <b v o "m "in VsV> 

lO CO M -»f COCOM COM-* H«)+ N »1 MM MM M Tf I/O 

co * r^ o co * co -*oo mm w ioj^coooo coo >o w m m m 

lO^O >0 Tj-O *0 "TJ-U1 ^ **■ ■*• ■«<■ ■* ■* ■*■<*■ MM MM -<i- CO ■<*- 



i I 

H.§sa.5- 



ail 







a i 

-c.3 



APPENDICES 



287 







ill 

£ rt u 



II iS 




II 



& § 



a a 



II 



££ £ 




3 . 

ST S3 



•as •: a 
a s s J > I 

<fl I* o rt O rt 



"O T3 cr 
o o « 

a - a - 1 



vS N 10 



O tx u-i O N «5 
O H CO CO N O 



3 3 4) O O O^ 3 aj^^3 ! 



S.S!" 






O M O 

I £ 9 






00 VO rt-^- 



I 8"3 

ft tr>y~> 






o & g J 

O co i— > S3 



tj-vOco 

^N ^N VO 



p m <u X /-< kS 



fs co CO Tj- Vp Th H 

NO H JNVO 



ao\« m coo ■* •*• co 

tX N VO 00'co * H 10 

0* CO CO M H 



CO^VO M M 

6 co fx ob V 



HNCt 

•ANN 


10 


10 Ao 


NVO 

tNVO 


O 

vo 


*s«n 

VO CO 


00 


VO 

VO 


VO 

vO 


m 




CO 
00 


CO 


00 


VO 

00 


O 

00 


0\ 0\lv 
VO vo <v, 




V0 

co 

vO 


in in 

moo 00 


VO 


0^0^ 


2R 


H 

00 


m co 


VO 


CO 


in 

VO 


VO 


CO 
00 


VO 




00 




Ov 
VO 


tNVO fN 


fs 


H VO N 
W N in 

1 


1 


^ 


"1 





co m 

1 


CO 


1 


VO 


1 


VO 

»n 


1 


1 


VO 


tN 


m 


m mvo 
mvo ^> 


m 


CO 

VO 


VO ^M 


1 


1 *> 

1 o> 


vo I 

o> 1 


vo 


000 

0\ Ov 


M 
Ov 


1 


00 


1 


o> 


1 


1 


Oi 


0\ 


O 
O 


OsOO On 


Ov 


On 
On 


« CO CO 

^oo 00 


R 


00 00 


xj-vO 





00 tv 


00 
m 


CO 

VO 




0\ 


00 


00 
vo 


Oi 
VO 


o> 




00 


u-) 


N O O 

00 CO 00 


O 

00 


CO 


CM-**- Oi 
w mvo 


i^- 
in 


*n in 

vo in 


M CO 
CO CO 


o> 

CO 


H OS 
CO <M 





O 


CO 
CO 


m 


m 


VO 
VO 


CO 

VO 


00 


O 
00 


tN 


M •* CO 

tvJ^vo 





VO 


(>. 

mtX J>, 


VO 


m 


in m 


CO 

m 


VO N 

in m 


VO 
10 


>n 


m 

m 


^. 




VO 


00 
VO 


0\ 


O 

00 




tNOO M 

t^ tv tv 


VO 


On 



^^^ £ ££ ^^ ^ ^^ £ £ £ £ 



£ £ £ £ £ 



£££ £ £ 



H vO vO 

KOi-i- 
fv mvo 



ZZZ, K ££ ^^ 5? ^^ 55 £ £ fc 



55 55 & 



"m 00 V» 00 V 



N M CO M 



it 



j '1 .is •?! 



sal 



>"d ca 

si & 



•31§ 
-••a 




wt^iiis|ft 



288 



APPENDICES 



li 








be >» 


><>. 








1 

"3 


1 














Jfl 


p p 

•a -a 


bo 






'3 . >> 


^J 






ed 








8 8 


"dti 







^1 1 
31 J 




>> 

p 


p 

•3 

p5 


o 




>. 




a a 


s 


>> 




IB 3 

u"« a 


*«L* 


"1 






a 




aT 


of oT 


"p 


p 




^)J2 cd 


uT 




Cd 




'3 




3 13 


33 


1 


'3 




,o ,o uT 


3 




a 




h 




cd cB 


rt c8 






cd cd |g 


cd cd p 


cd 


•3 




tj £ >> 

5 5 1 

000 

www 


a 

•3 

M 




3 p 

a a 


P s 
era 1 


>>^p 

a -p 

'rt 'Si 2 
000 


I 



w 


p 

'3 

11 



p p cr 

aaa^ 


p 

o" 


1/) 


a 


o\ •+ 


-* 


N u-> 


vO , 


CO M U") 


>o 


tN 


0) O VO 


On CO H 


tN 


VO M 


«J ^ -^ 


tN co M 


|N 


00 tN 


N 1 


lO 10 vo 


N 







■<*-vO ■* 


IN 


VO IN 


Q 


K 


CM CM 








M 0) M 




VN 




M 


M 


M 






vo 00 y-> 


<n 


n m 


« N 


u->vo 0\ 


rs 


p 


tN 


HNO 


^*— , 


000 




51 


.6 cm cm 


" 


w b 


6 6 


H N 6 


M 


CO 


b 
1 1 


b ii b 


00 N 


b b 









^f5 
p 3 p 


If 


bi> > 

POO 


"p 







1 1 ^. 




d,4> 


i^p 

P cd 


bo >» 

33 


3 




• a fe <i fe 


ft 


1— >►— i 1 — > 


l—><-> 


<:^^; 


*-) 







fel-ife 


)— )•— > 


<>-> 




. M M Th 


vp 


CO— *-— 


in u-) 


CO H VO 


«■> 


0\C0 


? tN 


CO H Ol 


CO U"> 


fN CM 


1 


a> 


8*S M *N 

A M 


bi 


V "9 


ij-vo 


6 Tj- H 


lO 


M H 


b m 


cooo 10 


VO VO 


b fs 




-* 




M M H 




M M 


1 .-" ' * 


M 




H 




p+j > w> 
■« a 

O fc <! 


4 

< 


2 Q 


a p 

cd cfl 




si 




■§ cd cd 


a> <p 
p p p 
p p p 
1 — >»—>•—» 


Oh O 


si 




e^i 00 10 










— Y^ 


>»,„/ 


^—^^ 




v_ v _' 






as 


u-> 


T*" ^ 


N N 


H 9 <N 


VO 


(N 


f p ? 


M >pp* 


UO 


CM H 




Q CO 00 VO 


CO 


tN tN 


N ^s 


M VO U-> 


CO 


00 


b b vo 


4j- co bv 


bi 


b co 




> 


VO 






^s o\vo 


->!- 


00 




H O M 


CO 


■* •* 


o>& 


>» 


£?vo CO tN 
In CO tN 





<+ ro 





tN ON<N 


00 


N 


N •* 0\ 


coO O 


N 


0\H 


rt'a 


"3 
>—> 


00 


00 vo 


CO v> 


tNOO tN 


tN 


00 


NN N 


00 Onoo 


tN 


tNVO 


























® 9 


a 

CO 

i-j 


>^0O CM tN 
In 00 In 





C^ fO 

VO vo 


m 

VO IN 


M N N 
00 OMn 


00 


00 


M CO CO 
tNfN VO 


vo IsN 


tN 


OVO 

00 IN 




d 


hi cs cm 10 


M 


tN O 




O Kts 


VO 





NO N 


«N00 C\ 


CM 


S 1 






">o io <o 


VO 




: ; 


tNVO O 






+ n n 




■*}• 


VO 1 


si 


S 
























S 


fl^ •* N CM 


u-> 


ov «o 


1 1 


OO CO O 


VO 





HO N 


N OWO 


^. 


S 1 


S 


o O On On 


on 


Ov O 


1 1 


Ov 0\ <J\ 


o» 


o-. 


00 00 IN 


0>00 00 


^ 


00 1 




>, 


r • M O O 

MH OO 00 OO 


00 


Ov O 


•<*- 0^ 


HOO K 


00 





*N O >0 


00 tN CO 


tN 


VO tN 


a; 


p 


tN 


Tj- >0 




00 *N *N 


vo 


00 


u->vo co 


tj- tj- m 


IO 


tN IO 


eg u 


»-> 

























d 


[Li VO VO V> 

^ in in in 


■0 


■* Tj- 


H U-) 


O 00 N 


00 


Ov 


O O N 


ON ■+ 


00 


O IN 


<D 0) 


tN 


In In 


vo m 


00 <N00 


<N 


<N 


tN tN >0 


vo vo vO 




00 <o 


s f 


>-» 
















































5 


3* 


tLi on O 00 


tN 


H N 


C\ * 


M OO O 


CO 





co m ^*- 


(7\ -* & 


00 


00 tN 




£ 


r^ ^ ^ 


tN 


VO vo 




OO tNOO 




00 


VO vo •* 


v " n " 1 




tNUO 


2 
b 




.0 (O <M 


CM 


a* 00 


M ^h 


Th <M In 


M 


vO 


00 m ci 


VOO 10 


00 





> 


-t-J M H 00 


IN 


00 00 
CO 


>* 

O CO 


■* * 0\ 







001 o> 


O HI CO 


«o 


CO 


'a 
U 




<u 




« 


M* 6\ 










M 


00" 


cR 


B 

C 

c 




? ^ ? 


£ 


^ ? 


f£ 


f&$ 


£ 


£ 


££ $ 


£££ 


i 


£f 


> 


00 vb Vs 


CO 


o> *« 


"ovVn 







"m 


Vo V 


nV» 

■* M CO 


"io 


m co 




1 


VO M VO 




00 %• 


00 m 


00 ■* 


co 



00 


°o°o °o 


O CO M 


%■ 


°O00 
tNtN 






in <o >o 


VO 


10 vo 


VO vo 


vo •* co 


T*- 


m 


NN N 


In tN tN 


^ 






fc £ fc 


£ 


(A c/i 


c/5c/5 


173C/5C/3 


w 


^ 


c/5c/5 w 


C/3C/3C/5 


% 


COvT) 


c 




00 on 


v o 


V> V. 


o"co 


00 «sV 


\f, 


"0 


*V< 00 


IN 00 In 


vo 


"cmV 




10 ro cs 








CS 




in 




M •^■ 


CO 




1 




























tN •<)- 00 





* M 


vO C\ 


CO M 00 




vb 


cooo co 


CO Ol lO 




0« O 










CO CO 






M 




CM M lO 


CO CO CO 














<i * • 




"aT 


















<v 









i. 





1 










( 

1 
< 

P 


i 
5 
> 

1 


ill ' '^ 

fllliili 


2 <n 

ill 1 


1 N 
•S « p 


• • 

p 

llsl 


•<3 

p 

cd 



s 


1 

2 i 

1 1 

1 


iS p 

00 "^ 

rt - 

111 1 




III 

■lis 

■5*9 Ph 


s 
g bO 


1 3 






^ fel 5 


h 


03 


DQ 




DQ 










1*1 



APPENDICES 



289 



£ £ £ ffi 



a ~ a - 

2S2 



a 8 5 



ffi £ 



W % X K £ 



O h w 



,£ 3 a 3 3 5^0 



3g § §| 9 



P. 

1—) 


SvV) 


rt 
SO 


a 


1—1 


S rt (i) c3 ™ 


N 






IN 


VO 






'— y— ' 


ON 


CO 


co 


H 


-* 


VO 





O. 


w 


CI 


O 


,— ' v 


H 


Ov 


o> 


co 


>> 








> 


bo 


bo 


si 




iS 




q£ 


fc 


3 


3 
< 


3 
< 



Q <1 < i=, Q 






•-2 B 

ft ft 



o S£ p 



In 


s 


CO 


O 

00 


O 

on 


VO 


00 

VO 


H Ti- 
N In 


CO 

CO 


CO 


In 

*N 


CO 


CO 


IN 


On 


VO 
In 


IN 


VO 


CO 
CO 


CO 
IN 


CO 

vo 


IN 

m 


IN 


10 

CO 


CO 

VO 


CO 

In 


In 
In 


10 


in On IN 

VO tUn 




CO 


CO 


vo 
In 


00 


M 

CO 


CO 
IN 


m 

00 


IN 

VO 


on 


CO 
In 


00 


vo 
m 


>n 


CO 


CO 


a 


VO 

CO 




CO 


1 


00 


|S 




VO 


1 


O 
VO 


1 


CO 


VO 


CO 
VO 


■>*■ 


ON 

CO 


IN 

VO 


m 

VO 


i 


1 


* 


10 

O 


CO 

00 


On 


00 


ON 

VO 


1 





1* 


o\ 


1 


VO 


1 


In 

ON 


H 


On 


CO 
O 


CO 




s 


o\ 


1 


1 


vO 


Th 
O 




vo 


•o 


00 

vo 


VO 
CO 


<£ 


in on co 


ON 

VO 


VO 


VO 
In 


CO 
|N 


VO 


•v. 

Oi 


in 

IN 


IN 
IN 


CO 


m 

On 


vo 

IN 


CO 


3 


On 

CO 


00 


In 


CO 


VO 


0\ 


CO 


co m co 


CO 

In 


In 


CO 
In 


IN 


In 


IN 





VO 


in 


On 

IN 


O 

CO 


In 


•* 

|N 


vO 

vO 


In 


vO 


CO 

VO 


VO 


N 


u-> 


Tt-00 CO 

vo Invo 


IN 


CO 

vo 


In 

in 


*N 

*N 


On 

VO 


VO 
00 


CO 


00 

VO 


0\ 
vo 


vO 
CO 


On 

IN 


In 


0O 


00 

CO 



£ £ £ £ £ 



ww^ w w 



www WW WW 



55 2Z?; co 

CI vO vO N Ov 



c/) en cfl 



fc £ 



25 2 

"O H 



290 



APPENDICES 



W 
« 

I 

a 

a 
u 





1 
















c 
























>» 






a 


>, 


>, 




>, 






u 










d 

're 

u 






1 


_d 
're 
u 


a 
1 


i 


d 

2 






re 

XI 

o 















>. 

'S 


_d 
're 




g 


a fi 9 

I 3 ' 3 


d 
"I 




a 


i 




_o 




i 






a _i) 


4T 


<lT 


oT 


of 


< aT 


of 


CO 




re 

a 




2 






'S3 


3 


2 


>>3 


3 


3 


A 


^ 












in cd 


rt 


CS 


d «s 


re 


"'re n 


re 


re 


d 












-; 3 





9 


'a s 


3 


>^9 • ^r 


3 


3 


'2 








T3 






if 


a< 


CT 


2 cr 


cr 


cr 


ct 











a^ 


1 


>> 
1 


03 


a 




a 


•3D 2.9 

a a a | 


(D 


V 


> 


■a I 
















l 


11 


s 
^ 

























11 


^ 


^ 


^ 


1 








HI ■ 

>»sH.a 




CO 


«* 


CM 


in fv. 


>o 


„. 


O N 


UO 


Mm vo -<f 


00 


VO 


co 


vo »n 


rt o re 




vo 


m 




U0 CM 


uo 


in 


00 OV 


CO 


M O^l M N 




00 


rv 


m m 


P g 










M M 






M 




M MM 


H 




H 













vp 


O 


fNOO 


9 


O 


O tt 


H 


°P ?*" ? 9 


CO 


p 


j, 


?x 




01 fl 
■Sri 







b 


O 


b b 





b 


^t-b 


b 


b b oh 


b 


b 


b 


1 


fijg 




_a 


si 


►-><!< 


4; m 


S bi>^bx) 


4J^ 

5 9 

Oh 


—< 


0.5? d w 

W3< hfl 




£ 3 


4a 

W 


0> '• ij 

a a « 

1— »•— »s 






■^-> 






mop 






p> UO 




t^ Ti- OO M 


„ 


v_^ 


m 


00 V_ Y— ' 


a!-^ 


ti 


vp 





ts 


CM C^ 


*P 


^s 


in m 


>n 


m in n ob 


H 


!*■ 


m 


b ^ 




b\ 


fN 


b 


M 


Cl 






m 










CO 


'3 


ti d 

CO O 

£2 


■a 


bi 

< 


1—5 


d 

Q 


cd a. 


d 
1-1 


<v 

Q 




> 

O 


1— >W <!•-, 


d 
re 


c 
re 
1—) 


>. 

<—> 


re a 

s 4 


£ 

.^ 


1 


fv 


vp 


M 


co CO 


10 


H 


vp ^f 


tJ- 


NO O CO 


m 


p 


fs 


h 9 




1 


ts 


JN 


CO 


A* b 


b 


UO 


00 M 


b 


00 b\ in n 


4h 


c* 


N 


m bi 




>■ 


T *° 


CM 




<+ ■* 






H lO 


CO 


in c< vn tN 


m 


CO 


■* 


H 


a) £ 


>> 

3 




Ov 


>n 


«+ 


<N. CM 


fN 


VO 


rhvo 


CO 


c?\ co >n cm 


m 


*s 


■* 


1 N 


># 


*n 


00 


10 


(X<N 


vo 


VO 


00 tN 


tN 


tVtN 0000 


fN 


>o 


00 


1 Ov 


■si 


H-) 




























































*fl 


d 


s - 


Jv. 


vo 


CO 


O M 


<N 


>n 


<♦ 0\ 





mvo >*• tJ- 


m 


■»i- 


CO 


1 *N 


a! 
1—) 


10 


*N, 


tN 


rsvo 


VO 


VO 


00 tN 


00 


<n(n co<n 


00 


tN 


tv 


1 00 




_d 


W 


tv. 


CM 


C\ 


00 


VO 


Cft 


O H 





COCO M <N 


m 


m 


vo 


*^ 1 






m 


vo 





O Tj- 


"* 


CO 


UOTt- 


->*• 


m ■* mo 




CO 


VO 


VO | 


S3 


§ 































X 






























cd 


W 




CO 


CO 


■* m 





u-> 


tv.00 


VO 




m 




VO 


CO | 


s 







Ol 





OvOO 








O^oo 


Ov 


OO O 01 


c* 


C7> 


O-i 


M | 







H 
















MM M 












ts 


W 


O 


tN 


„ 


t^ 


IN 


00 


O 00 


CO 


00 m vo 00 


o> 


VO 


Os 


VO Ov 


H 

9 
re S 

I 

H 


9 
1—) 




00 


In 


ov 


<v>£> 




*"* 


<n m 


*N 


vo vo (N <N 


m 


m 


tv. 


C7V tN 


i 

•-J 


W 


-*• 


M 


00 


N Tj- 


<N 


•>*• 


H O 


«o 


M iv. com 


O 


O 


H 


00 M 




tN. 


00 




00 VO 




10 


00 tN 


UO 


00 tv tx.00 


tx 


tN 


00 


*N 00 


13 


& 


00 


0\ 


U0 


H Tj" 


00 


tN 


in m 


CO 


VO M NH 


<n 


VO 


M 


O M 


£ 




<N 




00 


00 vo 


VO 


tNVO 


VO 


tN (N, tNOO 


VO 


VO 


00 


00 CO 


I 




0) 


00 


lO 


M O 


H 


r| 


O O 


^ 


co ■<*• m <M 


O 


UO 





O C\ 






00 




lO O 





tx 


c< O 


Oi 


N Ol M C< 


O 


CO 




CM CO 


c^ 








q\ 






■n 




m a> 




BO 






0) 


35 








10 












CO 


if 






a 




£ 


j£ 


w 


WW 


W 


W 


WW 


^ 


WW WW 


w 


w 


$ 


w w" 




vb 


"W 


vo 


V M \j- 


Vi 


V M 


W) 


V> 


"m vo *m V 


00 


v> 


\f. 


*bv "0 







CO 




CO 


>+ ■<*• 






M CO 


UO 


m co mm 








CM 


W 
































vo 







°0\0 


O 


CO 


O^tN 


m 




m 




CO 


co m 










* 


CO CO 


c< 


H 


•^ Tj- 




co co 


CO 


CO 


M 


<* i*- 






£ 


£ 


z 


inin 


^ 


£ 


iriin 


^ 


c/)c/) to to 


(A 


C/j 


to 


to to 






































CO 





•* ■* 






in m 


ts 


O 00 C» tN 


CO 


00 


Os 


m m 


W 






CO 








>o 


m 


■<i- 


mm co n 




-* 




CO 



































CO 


ID 


u-> 








00 00 


in 


Ov m O vo 


m 


"in 


CO 


M C« 














CO 


CO 


M M 


CO 


M <N M 


H 


H 




M 










1? 












0) 


















T3 































0) 












cr 










re 




1 tn 

1 « 

1 


,0 re 




ill 


■fi 


"o 


1(1 

iiii 




1 * | 

•1 Sis 1 s i 


1 


i 1 .2 

►32^3 1 






° 

a 1 




«0.b a S 

g^o re S Q 


8 


fll! 


p. 



11 %xv& 

^5 aj O *& ™ en 

§mw^«w 




i 


2S1^ O 

tW|h 2 








* 


^ ^ 




^ ^ 




^ fe; ^ as 




tyj 00 



APPENDICES 



291 



H> ***, *L> y*~> 3 3 u; 

M)r| i S S d 

I* 8 ! II I I 



.s.s.s 

M h h 
"O T3 T3 

111 



rO ,0,0. 

3 3 p 
crcr o 1 
<u <u <u 



a -a 



sg 



£££ £ 



5 .a 

as 

! !l 



II 



£ £ £ffi 




III 



I I? 



M 9 

6 M 



.0 ft a 




OD W) 

« 






3^0 




ts NO. 

o< VO vO 



bo ao 

3 3 



ass 



■2 &, 
w<! 






lO 00 M 



vp in 

fs H 

M VO 



N H 



NN CO 

MM U1 

MO ■* 





CO 


CO 


CO 


00 


10 




in m m 

t^vo ^s 


00 


O M 

VO 00 


00 


*n 


vO 


*M 


vo 


m 




<n 


CO 


Os N 


IT) 

00 


H 




CO 

<* 




CO 


VO 


00 


Cv covo 


CO 


1O00 


vo 
m 


•+ 


r>. 






00 
vo 


00 


? 




vo 


<N m 


m 


Os 
C* 


as 


1 


CO 

VO 


co 


VO 
m 


vo m O 

CO Tl- ■* 


*n 


CO Os 

«vO 


1 


CO 


Os 
CO 


^ O 

N vo 


Os 

CO 




Os 
vo 


VO 
CO 


CO 


txoo 
■* co 


00 
os 


Oi 


>+ 





Os 





00 


Os Os Os 


00 


M VO 

O Os 


1 


J 


O 




O Os 


8 


In 

O 


0^ 





in 
O 


vo in 

O 


lO 


m 


ON 

00 


vo 
(N 


VO 


OS 


OS 
VO 


m co os 
mvo in 


O 
10 


Osko 


xf 


m 


CO 

m 


IN *N 

in t-s 


Os 
m 


(x 


00 


in 


Os 


10 
in m 


VO 


t~> 


R 


CO 


CO 
00 


10 


*N 


M vo O 


VO 


in m 

<N0O 


OS 
vo 


co 


(n 


CO TT 
00 00 


jn 


in 

00 


00 




vO 


vO -*■ 
00 fN 


VO 


vo 


CO 
00 




Os 


VO 





co O m 
vo t^vo 


O 

vo 


CO Oi 
vO <N 


VO 

in 


01 

VO 


CO 

vo 


"+ co 

fxCO 


Os 
VO 


00 


00 


co 


00 
m 


O "T 
00 vO 



000 10 



W WW 
VO "co V» 



pq w WWW W WW 

*b "b\ ViV> V V vb "co 

M CJ C-) U"> -<t M 

O O OOO O OO 

O M COMtNOO T|-0> 

M CO H CON N WCO 



w 


W 


w* 


WW 


w 


w 


W 


w 


W 


WW 


10 


CO 


*N 


V)"m 


"■ M 


10 


N 


in 


CO 


m O 














































lO 




CO O 


CO 














■* 


w 5 


CO CO 


m 


■* 


2" 


co 


M 


M H 


co 


W3 


<f) 


CO CO 


(^ 


in 


C/3 


<S1 


in 


■J. J: 


"0 


V) 


N 


coco 


00 


Os 


V 


O 


Os 


rv(N 



% !? co J? 



CO CO CO CO 



COCO 
"coV) 



IN M m CO <N 

o 00 00 

in O co co N 




* «s 



co -^ &; 

w I ^ «* 

B||J| 



r 

1 *5 « ^^ 

"a ;§ j2 g s 
cflJ3o| 



1^ 3 

Pi < 






292 



APPENDICES 



i * 



.a3.s 

« d 3 



DPS 
cr cr a* 

<U <U D 

gag 
ill 



s -a 



™^ooo ^00000750 



cj o a 
Q « 



n 00 

M M I 



QS 



<! >-.OW 



Q 



»->►-»■< 



.C q X'M'o, 



►-><>—>>—> 



£ q 



h 4 



3 q^ "3 



II 



g ^ ^ £ Q Q 



£3 



rvt M ON t^ in OS 



CO \£> CO 



IT) tN. VO 



»>» tJ- vo <0 N M CO 
■**• 10 m vo *v *>. •<*■ 

I 



3 



<o 00 vo vo m <o 



M CO M 



.in >o m n o 

Qj M M H CI 



w w www w 



£ £ £ w 



£ £ w w £ w w w 



«N ts tN tv 



m m 



m m -<*- 



C/3 (/)WV) 
"o\ "o V» v m 



c« £ £ £ c/5 £ 



eo co co ca 



* CO 

CO ■*(• o\ 



APPENDIX II 
MEASURES OF HUMIDITY 





Relative Humidity. 


Air 
Tempera- 


100% 


60% 


20% 


ture. 




















Water 
content. 


Vapour 
pressure. 


Dew 
point. 


Wet 
bulb. 


Saturation 
deficit. 


Dew 
point. 


Wet 

bulb. 


Saturation 
deficit. 


°F. 

140 
135 


gr./cu.ft. 
58-5 
5i-4 


mb. 
199 

175 


°F. 

121 
116 


°F. 

123 
118 


gr./cu.ft. 

23-4 
20-6 


°F. 
84 
80 


°F. 
97 
93 


gr./cu.ft. 
46-8 
41-1 


130 
125 


45-i 
39'4 


153 
134 


in 
107 


1 14 
no 


18-0 
15-8 


76 

72 


90 
87 


36-1 
3i-5 


120 
115 


34-5 
30-1 


117 

IOI 


102 
98 


105 

IOI 


13-8 

12-0 


68 
64 


83 
80 


27-6 
24-1 


no 
105 


26-3 
23-0 


88 

76 


93 
88 


96 
92 


10-5 

9-2 


60 
56 


77 
74 


21 -o 
18-4 


100 
95 


20-0 

17-3 


65 
56 


84 
79 


88 
83 


8-o 

6-9 


52 
48 


7i 
68 


16-0 
13-8 


90 
85 


15-0 

12-9 


48 
41 


74 
69 


79 
75 


6-o 
5'2 


44 
40 


65 
62 


I2'0 

10-3 


80 

75 


I I-I 
9'5 


35 
30 


65 
60 


70 
66 


4.4 
38 


36 
32 


58 
55 


8-9 
7-6 


70 
65 


8-i 
6-9 


25 
21 


55 
5i 


62 
57 


3-2 
28 


28 
24 


52 
48 


6-5 
5'5 


60 
55 


5-8 
4'9 


18 
15 


46 
41 


53 

49 


2-3 

2-0 


20 
16 


45 

42 


4-6 
3-9 


50 
45 


4-1 
3'4 


12 
10 


36 
32 


44 
40 


16 
14 


12 
9 


38 
34 


3*3 

2-7 


40 
35 


2-9 
2-4 


8 

7 


28 
23 


35 
3i 


1-2 

I -o 


5 
1 


30 

27 


2-3 
»*9 


30 
25 


"*9 

i-5 


6 
5 


19 

14 


27 
22 


08 
o-6 


-3 

-7 


23 
19 


i'5 

1-2 


20 


1-2 


4 


9 


17 


9*5 


— 1 1 


15 


I O 



293 



APPENDIX III 
CONVERSION FACTORS 

Length 
i metre =39 37 inches = 3 281 ft. = 1-0936 yds. 

1 inch = 2 540 centimetres. 1 foot =0-3048 metres. 1 yard =0-9 144 metres. 
1 kilometre =0-62 14 miles. 1 mile = 1-609 km. 

Area 
1 sq. metre = 10-76 sq. ft. = 1-196 sq. yds. 1 sq. ft. =00929 sq. metre. 
1 hectare =2 -47 1 acres. 1 acre =0-4047 hectares. 
1 acre =4,840 sq. yds. 
1 sq. km. =0-3861 sq. mile. 1 sq. mile =2 -590 sq. km. 

Volume 
1 cu. cm. (c.c.)=oo6i cu. in. 1 cu. in. = 16-4 c.c. 
1 cu. metre =35 -3 cu. ft. — 1-31 cu. yds. 1 cu. yd. =0 -764 cu. metre. 
1 litre=6i-025 cu. ins. =0-88 quart. 1 qt. = 1-1365 litres. 
1 acre-inch (inch per acre) =3,630 cu. ft. 

Weight 
1 gram= 15-432 grains =0-0353 oz. 1 grain =0-0648 gram. 1 oz. =28-35 

grams. 
1 kilogram =2 2046 lbs. (av.). 1 lb. =0-4536 kg. 
1 ton (2,240 lbs.) = 1-064 metric tonne. 1 metric tonne (1,000 kg.) = 

2,204-6 lbs. = 0-984 ton. 
1 gm./cu. metre=o-437 grains/cu. ft. 1 grain/cu. ft. =2 29 gm./cu. metre. 

Energy 
1 kg./cal. (Cal.) = 1,000 gm./cal. =3-968 B.T.U. 1 B.T.U. = 252 gm./cal. 
1 gm./cal./cm. 2 =3-69 B.T.U./sq. ft. 
1 joule =0-24 gm./cal. 

Velocity 
1 ft. /sec. =0-682 m.p.h. =0305 metre/sec. 

1 metre /sec. =2-237 m.p.h. = 3-281 ft./sec. 1 m.p.h. =0-447 metre/sec. 
1 Knot = i-i5i5 m.p.h. 

Force 
1 kg./sq. metre=o-205 lb./sq. ft. 1 lb./sq. ft. =4-88 kg./sq. metre. 

Illumination 
1 lux=o-092g ft. -candles. 1 ft.-candle = 10-764 lux. 

Temperature 
6° C. = (32 + 1 -8(9) ° F. F F. = ( 5 */9 - 32) • C. 
T° Absolute =273+0° C. 

294 



REFERENCES 

Ailleret, P. (1946). "L'energie eolienne: sa valeur et la prospection des 
sites. Paris. 

Anon. (1927). "Effect of the gale of October 28th on electrical trans- 
mission lines." Meteor. Mag., London, 62, p. 277. 

Anon. (1944). "Rapid surveys of atmospheric pollution." London, Q,. J. R. 
Meteor. Soc, 70, p. 281. 

Ashworth, J. R. (1929). "The influence of smoke and hot gases from 
factory chimneys on rainfall." London, Q.J. R. Meteor. Soc, 55, p. 341. 

Bagnold, R. A. (1937). "The transport of sand by wind." London, 
Gto>£r. J., 89, p. 409. 

Bazett, H. C. (1948). "Human efficiency as affected by climate and its 
relation to social development." Advanc. Sci., London, 4, p. 348. 

Bedford, T. (1946). "Environmental warmth and its measurement." 
Medical Research Council, War Memo., no. 17. H.M.S.O. 

Beer, C. G. P. and L. B. Leopold (1947). "Meteorological factors in- 
fluencing air pollution in the Los Angeles area." Trans. Amer. Geoph. 
Union, 28, p. 1 73. 

Belasco, J. E. (1945). "The temperature characteristics of different 
classes of air over the British Isles in winter." London, Q. J. R. Meteor. 
Soc, 71, p. 351. 

Berenyi, D. (1948). "Effect of house-blocks of towns on the fall of tem- 
perature at night." Meteor. Mag., London, 77, p. 277. 

Bergeron, T. (1949). "The problem of artificial control of rainfall on the 
globe." Tellus, Stockholm, 1, p. 32. 

Bernard, M. M. (1932). "Formulas for rainfall intensities of long dura- 
tion." Trans. Amer. Soc Civ. Engrs., 96, p. 592. 

(1948). "A modernized flood forecasting service — an effective means 

of flood damage reduction." Engin. News-Record, New York, Dec. 9. 

Bilham, E. G. (1936). "Classification of heavy falls in short periods." 
London, Brit. Rain/., 75, p. 262. 

(1938). "The climate of the British Isles." London (Macmillan). 

(1947). "Here is the weather forecast." London (Golden Galley 

Press). 

Bilham, E. G., and R. F. Relf (1937). "The dynamics of large hail- 
stones." London, Q. J. R. Meteor. Soc, 63, p. 149. 

Billett, H. (19 14). "The South Wales tornado of October 27th, 19 13." 
London, Met. Off., Geoph. Mem., 2, no. n. 

Blair, T. A. (1937). "Weather elements." New York (Prentice-Hall). 

Bodroff, B. A. (1935). "Influence of shelterbelts on the micro-climate 
of the adjoining territory." Ail-Union Sci. Res. Inst, of Sylviculture 
and Improvement of Farmland by Forestation. Russian. 

Bonacina, L. C. W. (1927). "Snowfall in the British Isles during the half- 
century 1876-1925." London, Brit. Rain/., 1927 (1928), p. 260. 

Brooks, C. E. P. (1925). "The distribution of thunderstorms over the 
globe." London, Meteor. Off, Geoph. Mem., 3, no. 24. 

(1929). "The relation between the duration of bright sunshine 

registered by a Campbell-Stokes sunshine recorder and die estimated 
amount of cloud." London, Meteor. Off, Prof. Notes, no. 53. 

(1944). "The frequency distribution of hailstone sizes." London, 

Q. J. R. Meteor. Soc, 70, p. 227. 

295 



296 REFERENCES 

Brooks, C. E. P. (1946a). "Annual recurrences of weather; 'singularities \ " 

Weather, London, 1, pp. 107, 130. 
(1946b). "Climate and the deterioration of materials." London, 

Q,. J. R. Meteor. Soc, 72, p. 87. 
(1948). "Atmospheric pollution in Great Britain in relation to the 

burning of coal." London, Nat. Smoke Abatement Soc, Proc. Conf., 

Edinburgh, 1947 (repr.). 
(1949). "Frequencies of winds between given velocities." Meteor. 

Mag., London, 78, p. 33. 
Brooks, C. E. P., and N. Carruthers (1946). "The distribution of heavy 

rains in one and two hours." Water, London, June, p. 275. 
Brooks, C. E. P., and J. Glasspoole (1922). "The drought of 1921." 

London, Q,.J. R. Meteor. Soc, 48, p. 139. 

(1928). "British floods and droughts." London (Benn). 

Brooks, C. E. P., and G. L. Thorman (1928). "The distribution of mean 

annual maxima and minima of temperature over the globe." London, 

Meteor. Off., Geoph. Mem., 5, no. 44. 
Brooks, C. F. (1935). "Why the weather." New York (Harcourt, Brace 

& Co.). 
Brooks, C. F., and A. H. Thiessen (1937). "The meteorology of great 

floods in the eastern United States." New York, N.Y., Geogr. Rev., 27, 

p. 269. 
Brown, C. W., and W. O. J. Roberts (1937). "The areal frequency of 

tornadoes in the United States by counties, 1 880-1 931." J. Amer. 

Geoph. Union, 1935, Pt. I, p. 144. 
Brun, A. (1944). "Les locomotives trans-sahariennes et l'erosion sableuse." 

Comptes-Rendus, Paris, 218, p. 419. 
Brunt, D. (1947). "Some physical aspects of the heat balance of the 

human body." London, Proc. Phys. Soc, 59, p. 713. 
Bush, R. (1946). "Frost and the fruitgrower." London (Cassell), 2nd ed. 
Cabrera, N., and J. Hamon (1947). Sur l'oxydation de l'aluminium en 

atmosphere humide." Comptes-Rendus, Paris, 225, p. 59. 
Carruthers, N. (1943). "Variations in wind velocity near the ground." 

London, Q. J. R. Meteor. Soc, 69, p. 289. 
Carson, F. T. (1944). "Effect of humidity on physical properties of paper." 

U.S. Dept. Commerce, Circ. Nat. Bur. Standards, C.445. 
Castellani, Sir A. (1938). "Climate and acclimatisation." London 

(J. Bales). 
Chalmers, J. A. (1941). "Cloud and earth lightning flashes." Phil. Mag., 

32, p. 77- 

Chen-li (1938). "A study of room meteorological conditions in a cotton 

mill." Meteor. Mag., China, Oct., p. 95. 
Corless, R. (1930). "The gale of the 12th of January, 1930." London, 

Q,.J. R. Meteor. Soc, 56, p. 384. 
Cornthwaite, H. G. (1920). " Exposed steel temperatures in the tropics." 

Washington, D.C., U.S. Mon. Weath. Rev., 48, p. 403. 
Crabtree, J. (1948). "Rubber and the weather." New York, Bell Lab. 

Record, 26, p. 119. 
De Monts (1936). "Puissance destructive comparee des cyclones en 

Amerique et dans l'Ocean Indien." Paris, A. Phys. Globe, France 

d'Outre-Mer., 3, p. 157. 
Dight, F. H. (1934). "An analysis of warm spells in London from 1900- 

33, with special reference to the prevailing conditions of humidity." 
London, Meteor. Mag., 69, p. 109. 



REFERENCES 297 

Dobson, G. M. B. (1948). "Some meteorological aspects of atmospheric 

pollution." London, Q.J. R. Meteor. Soc, 74, p. 133. 
Doring, K. (1925). "Wind und Warrae bei der Berechnung hoher Schorn- 

steine aus Eisenbeton." Berlin. 
Durrell, A. T. (1947). "Severe squall at Kano, northern Nigeria." 

London, Meteor. Mag., 76, p. 206. 
Dufour, L. (n.d.). "Notes sur le probleme de la gelee nocturne." Brussels. 
Eades, H. W. (1945). "Rusting of cans in wooden and fibreboard boxes." 

Canada, Forest Products Lab., Vancouver Branch, Mimeo 100-V. 
Eliot, J. (1899). "Hailstorms in India during the period 1833-1897, with 

a discussion on their distribution." Calcutta, Ind. Met. Mem., 6, pt. 4. 
Faust, H. (1948). " Untersuchungen von Forstschaden hinsichtlich der 

Windstruktur bei einer Bo." Meteor. Rdsch., Heidelberg, 1, p. 290. 
Fitzmaurice, R. (1942). "Lighting, heating and ventilation." London, 

J. R. S. Arts, 90, p. 494. 
Fleming, R. (1930). "Wind stresses in buildings. With a chapter on 

earthquakes and earthquake resistance." New York. 
Fletcher, R. D. (1949). "The Donora smog disaster. A problem in 

atmospheric pollution." Weatherwise, 2, p. 56. 
Gallay, R., and P. Darbre (1948). "La lutte contre le gel au moyen de 

chaufferettes." Lausanne, Station fed. d'essais viticoles, arboricoles et 

de chimie agricole. Publ. no. 366. 
Geddes, A. (1946). "Planning and climate." London, Ass. Planning and 

Regional Reconstr., Rep. no. 38. 
Glasspoole, J. (1930). "The areas covered by intense and widespread falls 

of rain." London, Proc. Inst. Civ. Engin., 229, Pt. I, Paper no. 4738. 
Gold, E. (1936). "Wind in Britain. The Dines anemometer and some 

notable records during the last forty years." London, Q. J. R. Meteor. 

Soc., 62, p. 167. 
Golde, R. H. (1945). "Frequency of occurrence of lightning flashes to 

earth." London, Q. J. R. Meteor. Soc, 71, p. 89. 
(1946). "The frequency of occurrence and the distribution of light- 
ning flashes to transmission lines." London, Brit. Elec. Allied Industr. 

Res. Assn., Techn. Rep., S/T49b. 
Gottmann, J. (1942). "New facts and some reflections on the Sahara." 

New York, N.Y., Geogr. Rev., 32, p. 659. 
Grabham, G. W. (1921). "Some factors in thermal sanitation in the 

tropics." Cambridge, J. Hygiene, 19, p. 245. 
Grierson, R. (1941). "Degree days. Proposed basis for estimating the 

consumption of energy in space-heating installations." London, 

J. Inst. Heat. Ventil. Engrs., 8, p. 381. 
Grimmitt, H. W. (1945). "Ice storms and overhead lines." London, 

Q,. J. R. Meteor. Soc., 71, p. 286. 
Handisyde, C. C. (1947). "The climate of the home." London, Weather, 

2, 1947, p. 82. 
Harvey, W. F. (1930). "Experiments on radiant energy as a lethal agent." 

Ind. J. Medical Res., 17, no. 2. 
Hawke, E. L. (1944). "Thermal characteristics of a Hertfordshire frost- 
hollow." London, Q. J. R. Meteor. Soc, 70, p. 23. 

(1946). "Frost-hollows." London, Weather, 1, p. 41. 

Hebley, H. F. (1948). "Stability of the atmosphere and its influence on 

air pollution." Mech. Engin., New York, 70, p. 407. 
Hewson, E. W. (1945). "The meteorological control of atmospheric 

pollution by heavy industry." London, Q. J. R. Meteor. Soc, 71, 

p. 266. 



298 REFERENCES 

Hottel, H. C. (1941). "Artificial converters of solar energy." Washing- 
ton, Ann. Rep. Smiths. Inst., p. 151. 
Hrudicka, B. (1937-9). " Meteorologie im Dienste der Bautechnik." 

Leipzig, Z eitschr - &- Met., 54, pp. 37, 69; 55, pp. no, 289; 56, p. 248. 
Hudson, J. C. (1948). "The atmospheric corrosion of ferrous metals and 

its prevention." Gas World. 
Huntington, E. (1924). "Civilisation and climate." New Haven (Yale 

Univ. Press). 
Hurst, H. E., R. P. Black and Y. M. Simaika (1946). "The Nile Basin." 

Vol. 7. " The future conservation of the Nile." Min. Publ. Works, Cairo. 
Johnson, N. K., and E. L. Davtes (1927). "Some measurements of tem- 
peratures near the surface in various kinds of soils." London, Q.J. R. 

Meteor. Soc, 53, p. 45. 
Kanagy, J. R. (1947). "Adsorption of water vapour by untanned hides 

and various leathers at ioo° F." U.S. Dept. Commerce, Nat. Bur. 

Standards, J. Research, 38, p. 119. 
Kelly-Edwards, E. J. (1945). "Forestry notes for conservation officers," 

Pt. II. Salisbury, Rhodesia Agric. J., 42, p. 434. Repr. Salisbury Min. 

Agric. Lands. 
Kendrew, W. G. (1930). "Climate." Oxford (Clarendon Press). 
Kessler, O. W., and W. Kaempfert (1940). " Die Frostschadenverhutung." 

Berlin, Reichsamt. f. Wetterd., Wiss. Abh., 6, nr. 2. 
Kincer, J. B. (1937). "The national loss from windstorms, hail, floods, 

drought and other weather vagaries." Bull. Amer. Meteor. Soc, 18, 

p. 304. 
Kraus, E. B., and P. Squires (1947). "Experiments on the stimulation of 

clouds to produce rain." London, Nature, 159, p. 489. 
Lapworth, H. (1930). "Meteorology and water supply." London, 

Q. J. R. Meteor. Soc, 56, p. 271. 
Larose, P. (1947). "The effect of wind on the thermal resistance of clothing 

with special reference to the protection given by coverall fabrics of 

various permeabilities." Canadian J. Research, A., 25, p. 169. 
Lee, D. H. K. and H. Lemons. (1949). "Clothing for global man." Geogr. 

Rev., New York, 39, p. 181. 
Lemons, H. (1942a). "Hail as a factor in the regional climatology of the 

United States." New York, N.Y., Geogr. Rev., 32, p. 471. 
(1942b). "Hail in high and low latitudes." Bull. Amer. Meteor. Soc, 

23, p. 61. 
London, Admiralty (1938). "Admiralty weather manual." 
London, Air Ministry, Meteorological Office (1923). "The book of 

normals of meteorological elements for the British Isles." Section IV. 
(1940). "Hygrometric tables for the computation of relative humidity, 

vapour pressure, and dew point from readings of dry- and wet-bulb 

thermometers exposed in Stevenson screens." M.O.265. 
London, Royal Meteorological Society (1945). "Weather and electric 

power systems." London, Q. J. R. Meteor. Soc, 71, p. 283. 
Madison, Wis., U.S. Dept. Agriculture, Forest Products Laboratory 

(1944). "Study of temperature and moisture content in wood aircraft 

wings in different climates." Rep. no. 1597. 
Manley, G. (1939). "On the occurrence of snow-cover in Great Britain." 

London, Q.J. R. Meteor. Soc, 65, p. 2. 
(1948). "Records of snow cover in Scottish mountains." London, 

Meteor. Mag., 77, p. 270. 
Markham, S. F. (1942). "Climate and the energy of nations." London 

(Milford, Oxford Univ. Press). 



REFERENCES 299 

Marsh, A. (1947). "Smoke, the problem of coal in the atmosphere." 

London (Faber & Faber). 
Marshall, W. A. L. (1948). "London temperatures." London, Meteor. 

Mag., 77, p. 54. 
Meetham, A. R. (1948). "The measurement of atmospheric pollution." 

London, Nat. Smoke Abatement S., Proc. Con/., Edinburgh, 1947. Repr. 
Middleton, C. A. (1937). "Structural design in the tropics. Failure of a 

gasometer." Civil Engng., 32, p. 47. 
Midowicz, W. (1948). "An old method of obtaining water from the air 

in dry localities." London, Meteor. Mag., 77, p. 18. 
Morel, Ch. (1939). "La foudre et les installations electriques interieures." 

Bull. A. Suisse elec, no. 1. 
Muller-Hillebrand, D. (1937). "Beitrag zur Frage der Existenz besonders 

blitzgefahrdeter Stellen." Bucuresti, Rumdnisches Energie-Inst., nr. 138. 
Nageli, W. (1943). " Untersuchungen uber die Windverhaltnisse im 

Bereich von Windschutzstreifen." Zurich, Mitt. Schweiz. Anst. fbrstl. 
Versuchswesen, 23, p. 223. 

(1946). "Weitere Untersuchungen ..." idem, 24, p. 659. 

New York, American Society of Heating and Ventilating Engineers 

(1944). "Effective temperature chart showing normal scale of effective 

temperature. Applicable to inhabitants of the U.S. under certain 

conditions." Guide, p. 61. 
New York, Institution of Radio Engineers (1943). "Lightning striking 

frequencies for various heights." Proc., p. 79. 
New York, Metropolitan Insurance Co. (1948). "Death toll from 

lightning." Bull. Amer. Meteor. Soc, 29, p. 333. 
Oxford, Institute of Agricultural Engineering (1926). "A report on 

the use of windmills for the generation of electricity." Bull., no. 1. 

Oxford (Clarendon Press). 
Parde, M. (1946). "Etudes critiques sur la hauteur des crues." Geneva, 

Rev. Etude Calam., 9, p. 1. 
Pardy, A. A. (1946). "Wind-breaks and shelter belts." Salisbury, Rhodesia 

Agric. J., 43, p. 131, repr. as Bull., no. 1346, of Min. Agric. Lands. 
Parker, A. (1945). "Coal in relation to atmospheric pollution." London, 

Nature, 155, p. 682. 
Porter, L. C. (1934). "Light transmission through snow." New York, 

Trans. Ilium. Engin. Soc., p. 93. 
Ramdas, L. A., and R. K. Dravid (1936). "Soil temperatures in relation 

to other factors controlling the disposal of solar radiation at the earth's 

surface." Proc. Indian Acad. Sci., Calcutta, 2, p. 131. 
Reitschels, H. (1934). "Leitfaden der Heiz-und Luftungstechnik." 10 

Aufl. Berlin. (Ch. VIII, "The physical laws of moist air," by F. 

Merkel. Ch. IX, "Meteorological and climatic bases of heating and 

ventilation technique," by F. Bradtke.) 
Richards, B. D. (1944). "Flood estimation and control." London 

(Chapman and Hall). 
Roberts, B. (1943). "The study of man's reaction to a polar climate." 

Cambridge, Polar Record, 4, p. 63. 
Rowen, J. W., and D. Gaglierdi (1947). "Properties of water-repellant 

fabrics." Washington, Nat. Bur. Standards, J. Research, 38, p. 103. 
Ruby, Col. (1938). "Die neuen Hagelbekampfungsverfahren." Strass- 

bourg, Ann. Phys. Globe, T.i, Meteorologie, Annexe. 
Schaffer, R. J. (1938-9). "Atmospheric pollution in its bearing on the 

weathering of stone and similar building materials." London, Dept. 

Sci. Industr. Research, Invest. Atmosph. Pollution, 25, p. 1 1 . 



300 REFERENCES 

Schiller, P. (1944). "An analysis of the load on a modern electric-supply 

system." London, J. Inst. Elec. Engin., 91, Pt. II, p. 433. 
Schonland, B. F. J. (1938). "The lightning discharge." Oxford (Clarendon 

Press). 
Schulze, W. M. H. (1941). "Tropical climates and communication tech- 
niques." Elektr. Nachrichtentechnik, 18, p. 134. 
Sherlock, R. H., and E. A. Stalker (1940). "The control of gases in the 

wake of smoke stacks." Mech. Engin., June, 1940. 
Simpson, G. C. (1929). "Lightning." London, J. Inst. Elec. Engin., 67, 

p. 1269. 
Smith, T., and E. D. Brown (1944). "The natural lighting of houses and 

flats, with graded daylight factor tables." London, Nat. Phys. Lab. 
Snow, D., M. H. G. Crichton and N. C. Wright (1944-5). "Mould 

deterioration of feeding stuffs in relation to humidity of storage." 

Ann. Appl. Biol., 31, p. 102; 32, p. 40. 
Snow, D., J. A. B. Smith and N. G. Wright (1,944). "The manufacture 

and storing of cereal feeding stuffs with incorporated non-protein 

nitrogen compounds." J. Agric. Soc, 35, p. 65. 
Snow, D., and N. C. Wright (1944). "The respiration rate and loss of 

dry matter from stored bran." J. Agric. Sc, 35, p. 126. 
Stagg, J. M. (1950). "Solar radiation at Kew Observatory." London, 

Meteor. Office, Geoph. Mem., 86. 
Stone, R. G. (1940). "The distribution of the average depth of snow on 

ground in New York and New England." T. Amer. Geoph. Union, p. 672. 
(1941)- "Health in tropical climates." Washington, D.C., Dept. 

Agric, Tearb. Agric, p. 247. 
(1944)' "The average length of the season with snow-cover of various 

depths in New England." T. Amer. Geoph. Union, p. 874. 
Sutton, O. G. (1947). "Smoke from factory chimneys." London, Weather, 

2, p. 249. 
Tannehill, I. R. (1944). "Hurricanes. Their nature and history, par- 
ticularly those of the West Indies and the southern coasts of the United 

States." 5 ed. Princetown (Univ. Press). 
Telkes, M. (1947). "Solar house-heating — a problem of heat storage." 

Heat. Vent., New York, 44, p. 68. 
ter Linden, A. J. (1938). "Das Sommerklima in Gebauden." Gesundshtng., 

61, p. 522. 
Thelm, W. (1933). " Massgebliche Schlagregenwirkungen in der Bau- 

technik." Ann. Hydr., Berlin, 61, p. 196. 
Thomas, A. M. (1930). "Memorandum on high wind pressures on tall 

structures. A study of their magnitude and probable frequency." 

Brit. Elec. Allied Industr. Research Assn., Tech. Rep. F/T 42. 
Trombe, F. (1948). "Utilization of solar energy." Research, London, 1, 

P. 393- 
Vernon, W. H. J. (1935). "A laboratory study of atmospheric corrosion 

of metals." Pts. 2, 3, Trans. Faraday Soc, 31, p. 1668. 
(1945)' " Controlling factors in atmospheric and immersed corrosion." 

London, J. Sci. Instrum., 22, p. 226. 
Wallace, W. K. (1949). "Snow clearance on railways." Journ. Glaciol., 1, 

p. 300. 
Ward, Robert DeCourcy. (1925). "The climates of the United 

States. Boston (Ginn & Go.) 
Washington, U.S. Bureau of Standards. "Effect of solar radiation upon 

balloons." Technol. Paper, no. 128, p. 115. 



REFERENCES 301 

Washington, U.S. Dept. Commerce, National Bureau of Standards 

(1945). "Code for protection against lightning." Parts I, II, III. 

Handbook, H. 40. 
Weeks, J. R. (1942). "Degree day temperature manual." Baltimore, Md. 

State Weather Service. 
Wegener, A. (191 7). "Wind- und Wasserhosen in Europa." Braunschweig 

(F. Vieweg). 
Wright, H. L. (1932). "Observations of smoke particles and condensation 

nuclei at Kew Observatory." London, Meteor. Off., Geoph. Mem., 6, 

no. 57. 
Wright, N. C. (1940). "The storage of artificially dried grass." J. Agric. 

Sc. y 31, p. 194. 
Yarnell, D. L. (1935). "Rainfall intensity-frequency data." Washington, 

Dept. ^Vgric, Misc. Publ., 204. 



INDEX 



Aachen, 281 

Aberdeen, 280 

Abyssinia, 141, 289 

Acclimatisation, 31 

Accra, 139, 290 

Accumulated temperature, 235 

Addis Ababa, 289 

Adelaide, 73, 113, 114, 291 

Aden, 283 

Adrets, 107 

Afghanistan, 283 

Africa, 22, 26, 109, 289 

East, 153, 164 

South, 72, 142, 291 

West, 139, 221 
Ailleret, P., 36, 295 
Air conditioning, 244, 247 

conductivity, 74, 82 

cooling power, 30, 245 

density, 163 

drainage, 266 

ionisation, 262 

mass, 59, 60, 64 

moisture content, 17, 293 

movement, 245 

moving, energy, 272 

purity, 78 
Aircraft wing, temperature, 1 74 
Ajaccio, 281 

Alaska, 115, jy^ 166^2 86_ 
Albania, 280 
Albany, W. A., 114 
Alexandria, 71, 109, 289 
Algiers, 289 
Alice Springs, 114, 291 
Allahabad, 71, 284 
Alps, climate, 107 
Aluminium, oxidation, 184 

paint, 178 
American Society of Heating and Ven- 
tilating Engineers, 244, 299 
Amman, 71, 285 
Ammonium sulphate, 189 
Amoy, 136 

Andes, 149, 150, 156, 163 
Anemometer, 87, 100 
Angola, 289 
Ankara, 71, 285 
Anticyclone, 96 
Antofagasta, 288 
Appendicitis, 152 
Arabia, 159, 283 
Archangel, 105, 283 
Arctic climate, 165 



Arequipa, 164 
Argentine, 72, 148, 221, 288 
Arica, 288 
Artesian wells, 1 1 3 
Ascension Island, 292 
Ashworth, J. R., 258, 295 
Asia, 31, 130, 283 

Minor, 109 

monsoons, 131 

sea ice, 51 
Astrakhan, 283 
Asuncion, 289 
Athens, 69, 281 
Atmospheric Pollution, see Pollution, 

Atmospheric 
Atomic energy and weather control 

273 
Auckland, 73, 292 
Australia, climate, 30, in, 144, 291 

cyclones, 219, 221, 226 

rainfall, 113 

sunshine, 73 
Austria, 68, 280 
Azizia, 160 
Azores, 292 
Azov, Sea of, ice, 51 

Bacteria, growth of, 180, 185 

Baghdad, 284 

Bagnold, R. A., 162, 295 

Baguios, 226 

Bahamas, 146, 287 

Bahrein, 283 

Bai-u, 138 

Balbao Heights, 72, 287 

Balearic Isles, 282 

Balloon, temperature in, 1 74 

Baltic ice, 50, 54 

Baluchistan, 283 

Bangalore, 71, 284 

Bangkok, 285 

Barbados, 287 

Barcelona, 70, 282 

Barnaoul, 105, 285 

Basra, 284 

Batavia, 285 

Bathurst, Gambia, 139, 290 

Bazett, H. C, 248, 295 

Bedford, T., 248, 295 

Beer, C. G. P., 198, 295 

Beira, 290 

Beirut, 285 

Belasco, J. E., 76, 295 

Belfast, 280 

Belgium, 68, 280 



3O4 INDEX 

Belgrade, 70, 283 

Belize, 287 

Benghazi, 290 

Berbera, 289 

Berenyi, D., 67, 295 

Bergen, 282 

Bergeron, T., 260, 295 

Berg winds, 142 

Berlin, 69, 103, 104, 281 

Bermuda, 73, 148, 287 

Bernard, M. M., 204, 295 

Berne, 70, 282 

Bilham, E. G., 103, 203, 211, 295 

Billett, H., 221, 295 

Birmingham, 68, 192, 280 

Black, R. P., 298 

Black-bulb thermometer, 175 

Black Sea, bora, no 

ice, 51, 55 
Blair, T. A., 220, 295 
Blizzards, 43, 100, in, 116, 120, 125 
Blood-rain, 229 
Bodroff, B. A., 270, 271, 295 
Body-temperature, 21, 246 
Bogota, 164, 288 
Bohorok, 156 
Bolivia, 73, 288 
Bolobo, 139 
Bombay, 71, 284 
Bonacina, L. C. W., 101, 295 
Bora, no 
Bordeaux, 281 
Borneo, 283 
Boston, Mass., 286 
Bourke, N.S.W., 291 
Bradtke, F., 235 
Brazil, 26, 154, 288 
Breslau, 281 
Brisbane, 73, 291 
Britain, 

atmospheric pollution, 193 

climate, 30, 97, 280 

degree-days, 235 

droughts, 99 

floods, 100 

fog, 42 

frost-days, 81 

gales, 100 

glazed frost, 47, 213 

hail, 211 

ice-days, 81 

rainfall, 84, 99, 203 

snow, 212 
cover, 44, 99 

snowstorms, 100 

sunshine, 68 

temperature, 97 

thunderstorms, 205 

tornadoes, 221 

weather recurrences, 102 

wet-bulb, 27 



British Guiana, 73, 156, 288 

Honduras, 287 

Somaliland, 289 
Brno, 68 

Brooks, C. E. P., 36, 67, 100, 101, 104, 
172, 176, 203, 204, 211, 216, 242, 
295> 296 
Brooks, C. F., 120, 121,216,220,228,296 
Broome, 114, 291 
Brown, C. W., 218, 220, 296 
Brown, E. D., 250, 300 
Bruckner cycle, 103 
Brun, A., 296 
Brunn, 281 
Brunt, D., 245, 296 
Brussels, 68, 280 
Bucharest, 70, 282 
Budapest, 69, 282 
Buenos Aires, 72, 149, 288 
Building, 

interruption by frost, 81 
rain, 83 

on frozen ground, 167 
Buildings, 

air conditioning, 244 

corrosion of, 189 

damage by atmospheric pollution, 79 
lightning, 215 

effect on lighting, 78 

glare from, 59 

heating, 233 

by solar radiation, 57 

humidity in, 247 

lighting, 250 

loss of heat from, 67, 243 

oscillation period, 90 

suitability for climate, 56, 58 

temperature in, 58, 240 

wind-pressure on, 88, 90 
Bulgaria, 68, 280 
Buran, 106 
Burma, 22, 131, 283 
Bush, R., 264, 267, 296 
Bushire, 284 

Cabrera, N., 184, 296 

Cacimbo, 152 

Cadiz, 282 

Cairo, 59, 71, 109, 140, 289 

Calabar, 139 

Calcutta, 71, 284 

California, 

climate, 30, 128 

fog, 198 

rain, 201 

sunshine, 128 
Cameroons, 289 
Canada, 

Arctic, 166 

climate, 30, 1 15, 1 19, 1 24, 1 26, 127, 286 

debacle, 52 



INDEX 



305 



Canada — continued 

hail, 211 

heat-stroke, 22 

sea ice, 48, 53 

snow cover, 43 

sunshine, 72 
Canary Islands, 292 
Canberra, 73, 114, 291 
Candia, 281 
Cans, rusting of, 183 
Canton, 283 
Cape Spartel, 290 
Cape Town, 72, 291 
Caracas, 73, 289 
Cardiff, 68, 195, 280 
Carnarvon, W. A., 114 
Caroline Islands, 292 
Carruthers, N., 203, 204, 225, 296 
Carson, F. T., 181, 296 
Castellani, Sir Aldo, 151, 161, 176, 256, 

296 
Celebes, 283 
Cement, 

conductivity of heat, 66 

protection from frost, 81 
Central America, 146, 287 
Ceylon, 27, 157, 283 
Chalmers, J. A., 215, 296 
Charbin, 135, 285 
Charleston, 72, 123, 286 
Charlottetown, 286 
Chemical reactions 

and heat, 179 
humidity, 184 
Chemulpo, 71, 135, 285 
Chen-li, 32, 296 
Cherbourg, 281 
Cherrapunji, 131 
Chesterfield, Canada, 286 
Cheyenne, 117 
Chicago, 72, 119, 286 
Chihili, 161 

Chile, 73, 129, 149, 288 
Chills, 152, 167 
Chimneys, 

pollution by, 92, 93 

smoke from, 88, 188 

wind pressure on, 89 
China, 31, 70, 134, 283 
Chinook, 125 

Christchurch, N.Z., 73, 292 
Chubascos, 146 
Chungking, 136, 283 
Climate, 

Arctic, 165 

characterisation of, 278 

control, 233, 258 

classification, 17 

cyclonic-temperate, 20, 33, 97 

desert, 159 

deterioration due to, 20, 171 



Climate — continued 

ideal, 30 

insolation, 18 

Mediterranean, 20, 33, 112 

monsoon, 130 

mountain, 18, 61, 163 

polar, 18 

subtropical, 130 

suitability for industry, 31 

tables, 277 

temperate, 31, 95 

tropical, 30, 151 

variability, 30 

winter frost and snow, 21 
Clo, 254 
Cloches, 265 
Cloth, 

bleaching by UV rays, 176 

colour and temperature, 176 

insulating power, 254 

permeability, 254, 255 
Clothing, 31, 253 

Arctic, 167 

desert, 162 

tropical, 31 
Cloud, 

artificial, 265 

dissipation, 260 

effect on light, 78 

on temperature, 57, 265 
Cloudbursts, 128 
Coal 

and atmospheric pollution, 188, 192. 
196 

consumption, 244 
Cocos-Keeling Islands, 157 
Cold front, 96, 221 
Cold wave, 116, 120 
Colour, 

effect on temperature, 59, 176 

of headgear, 256 
Colombo, 158, 283 
Columbia, 288 
Comfort, 23, 26, 233, 246 
Condensation, 187, 200 
Cone of protection (lightning), 263 
Confectionery, 32, 79 
Congo, 152, 289 
Conjunctivitis, 256 
Constantinople, 283 
Conversion factors, 294 
Cooktown, 114 
Cooling of air, 248 

power, 23, 27 
Copenhagen, 68, 281 
Cordoba, Argentine, 72, 288 
Cork, Eire, 280 
Corless, R., 272, 296 
Cornthwaite, H. G., 175, 176, 296 
Corrosion, 180, 188 

by salt, 200 



306 INDEX 



Corsica, 281 

Cosmetics, 79 

Costa Rica, 287 

Cotton industry, 31, 32, 123 

Crabtree,J., 179,296 

Crete, 281 

Crichton, M. H. G., 180, 183, 300 

Cristobal, 287 

Cuba, 146, 287 

Cumulative curve, 237 

Cyclones, 

tropical, 142, 157, 219, 225 

Australia, 112, 219 

India, 133, 219 

see also Baguios, Hurricanes, Ty- 
phoons, Willy-willies 
Cyprus, 280 
Czecho-Slovakia, 68, 281 

Dakar, 139, 289 
Damage by 

cyclones, 112, 122, 133, 137, 147, 227 

glazed frost, 213 

hail, 211 

lightning, 263 

tornadoes, 218 
Danube, freezing, 51 
Danzig, 69, 282 
Darbre, P., 267, 297 
Dar-es-Salaam, 291 
Darwin, 112, 114, 291 
Davies, E. L., 59, 298 
Dawson, 286 
Day, length of, 76 
Daylight, 76, 250 
"Dead air," 253 
Death Valley, 127, 160 
Deaths by 

flood, 208 

hurricane, 227 

lightning, 214 

tornadoes, 218 
Debacle, 52 
Deforestation, 258 
Degree-days, degree-hours, 234 
Dehumidification, 182, 187, 248 
Delhi, 284 
De Monts, 227, 296 
Denmark, 68, 281 
Density, air, 163 
Denver, 72, 287 
Depression, 

barometric, 96 
energy, 272 

industrial, 31 

secondary, 222 
Deserts, 18, 159 
Deterioration, 20, 182 

index of, 185 
Dew, 161, 187 
Dew-point, 17, 293 



Dight, F. H., 248, 296 

Dobson, G. M. B., 196, 297 

Donora, 197 

Doring, K., 90, 297 

Dor pat, 281 

Dorrell, A. T., 222, 297 

Dover, 280 

"Downwash" of smoke, 93 

Dravid, R. K., 176, 299 

Drought, 

Africa, 154 

Britain, 99 

U.S.A., 122 
"Dry ice," 259 
Duala, 139, 289 
Dublin, 68, 280 
Duchemin, 90 
Dufour, L., 269, 297 
Dufton, A. F., 250 
Dunedin, 292 
Dunn's equation, 179 
Durazzo, 280 
Durban, 291 
Dust, 144, 161 

Bowl, 228 

storms, 117, 228 
Dutch Harbour, 286 

Eades, H. W., 182, 187, 297 

Eala, 289 

Earth current (lightning), 263 

East Africa, 153, 164 

East Indies, 156 

East London (S. Africa), 291 

Ecuador, 73, 288 

Eddies, 78, 85 

diameter, 86 
Edinburgh, 68, 280 
Edmonton, Canada, 72, 286 
Efficiency (human), 21, 30, 31, 248 
Egypt, 71, 109, 140, 256, 289 
Electric apparatus 

and lightning, 263 

effect of heat, 181 
humidity, 181 
salt nuclei, 200 
Electric niagara, 262 
Eliot, J., 211, 297 
Elizabethville, 289 
Eneseisk, 105, 285 
England and Wales, see Britain 
Entebbe, 291 
Equation of time, 64 
Eritrea, 289 
Erythema, 256 
Estonia, 281 

ice, 55 

sunshine, 68 
Europe, 

Central, 103 

climate, 30, 103, 280 



Europe — continued 

Eastern, 105 

North-west, 103 

sea ice, 50 

snow cover, 43 
European monsoon, 104 
Evaporation, 30, 41, 100 

and water supply, 39 

loss of body heat by, 22, 30 
Exfoliation, 80 
Expansion, coefficient of, 66 
Eye troubles, 144, 161, 256 

Factories, siting of, 56 
Falkland Islands, 129, 289 
Falmouth, 280 
Fan, anti-frost, 266 

electric, 245 
Faroes, 281 
Faust, H., 221, 297 
Fiji Islands, 292 
Finland, 51, 54, 68, 105, 281 
Fitzmaurice, R., 244, 250, 297 
Fleming, R., 90, 91, 297 
Fletcher, R. D., 197, 297 
Floods, 165, 207 

prediction, 208 

Britain, 100 

U.S.A., 121, 123, 208 
Flushing, 104, 281 
Fog, 42, 79,93, 196 

dispersal, 260 

Donora, 197 

London, 42 

Meuse Valley, 94 

North America, 128 

overhead, 252 
Fohn, 108 
Foot-candle, 76, 251 
Forecasting, 103 

flood, 204 

frost, 268 
Forel, Charles, 264 
Forest fires, 215 
Formosa, 70, 137, 284 
France, 69, 281 
Frankfurt-am-Main, 281 
French Africa, 289 
Frequency distribution, 236 
Frost, 79, 98 

and building period, 81 

and road surface, 48, 80 

black, 264 

damage to walls, 80 

forecasting, 268 

frequency, 80 

glazed, 47, 120, 213 

hollows, 80, 266 

penetration, 81 

protection, 264 

white, 264 

u* 



INDEX 

Fruit, 119, 123, 126 
Fuel consumption, 235 
Funchal, 292 
Fusan, 135 

Gaglierdi, D., 255, 299 

Gallay, R., 267, 297 

Galveston, 287 

Gambia, 290 

Gasometer, wind pressure, 89 

Geddes, A., 56, 297 

Geneva, 282 

Genoa, 69, 282 

Germany, 69, 281 

Ghibli, ghiblitis, 161 

Gibraltar, no, 281 

Glare, 256 

Glasgow, pollution, 194 

Glasspoole, J., 100, 207, 296, 297 

Glazed frost, 47, 120, 213 

Glycerine, flow of, 180 

Gold, E., 86, 297 

Gold Coast, 290 

Golde, R. H., 215, 263, 297 

Gottmann, J., 161, 297 

Grabham, G. W., 176, 297 

Great Britain, see Britain 

Great Plains, 117, 124 

Great Lakes, 34 

Greece, 69, 281 

Greenland, 166 

Grenada (W. Indies), 147 

Grierson, R., 234, 235, 297 

Grimmitt, H. W., 213, 297 

Guatemala, 287 

Guiana, 155 

Gust, 85 

maximum, 223 
Gustiness, 86 
Guyaquil, 288 

Haboob, 141, 161, 229 

Haifa, 285 

Hail, 112, 117, 143, 210, 211 

insurance, 212 

prevention, 260 

structure, 261 
Hair hygrometer, 181 
Haiti, 287 
Hakodate, 137 
Halifax, N.S., 286 
Hamburg, 69, 281 
Hamon, J., 184, 296 
Handisyde, C. C., 243, 297 
Hankow, 136 
Haparanda, 282 
Harmattan, 140 
Harvey, W. F., 176, 179, 297 
Havana, 146, 147, 287 
Hawaii, 292 
Hawke, E. L., 80, 297 



307 



3 o8 

Headgear, 256 
Health, 21 

and fog, 188 

in tropics, 151, 157 
Heat, 

conduction, 21, 75 

conservation, 243 

oedema, 152 

prostration, 121 

radiation, 21 

stroke, 22, 245 

waves, 121 
Heaters, anti-frost, 267 
Heating, 233, 242 
Hebley, H. F., 198, 297 
Hebron, 286 
Helsinki, 68, 281 
Hewson, E. W., 198, 297 
Hobart, Tasmania, 73, 114, 292 
Holland, 69, 281 
Hong Kong; 70, 136, 137, 283 
Honolulu, 292 
Hottel, H. C, 38, 298 
Houses of Parliament, 189 
Houses, siting of, 56, 58 
Hrudicka, B., 81, 298 
Hsingking, 135 
Hudson, J. C, 189, 298 
Hudson Bay, 49, 166 
Humidification, 32, 248 
Humidity, 277 

and temperature, 184 

critical, 181, 184 

effect on corrosion, 181, 184 

low, 160 

measures of, 17, 293 

optimum, 245 
Hungary, 69, 282 
Huntington, E., 21, 298 
Hurricanes, 118, 122, 

226 

Hurst, H. E., 141, 298 
Hydro-electricity, 33, 99, 108, 165 
Hygrometer, hair, 181 
Hygrometric tables, 277 

Ice, 

bridge, 49 

days, 80 

drift, 49 

Newfoundland, 49 

Saints, 104 

sea, 48 

storm, 47, 120, 213 
Iceland, 73, 166, 282 
Ichang, 136 
India, 

climate, 26, 31, 131, 284 

cyclones, 133, 226 

hail, 2 1 1 

heavy rain, 201 



INDEX 



24, 147, 219, 



I ndia — continued 

health, 22 

sunshine, 71 
Indian summer, 117, 150 
Indo-China, 26, 71, 131, 284 
Innsbruck, 68, 280 
Insalah, 160, 289 
Insolation, 18, 256 

protection against, 256 
Institution of Electrical Engineers, 251 
Insulation, 

electrical, 181, 200 

thermal, 253 
Insurance, 

hail, 212 

tornado, 220 
Inversion, 93, 198, 199, 265 
Iran, m, 284 
Iraq, 22, III, 284 
Ireland, 68, 280 
Irkutsk, 285 

Iron, corrosion of, 180, 184 
Irrigation, 126, 128 
Isokeraunic level, 215 
Istanbul, 283 
Italy, 69, 282 

Jacksonville, 123 
Jamaica, 73, 147, 288 
January thaw (U.S.A.), 117 
Japan, 71, 137, 284 
Java, 27, 156, 285 
Jerusalem, 285 
Jibouti, 290 
Jidda, 283 
Jinsen, 71, 135, 285 
Johannesburg, 291 
Johnson, Sir Nelson, 59, 298 

Kabul, 283 

Kaduna, 290 

Kaempfert, W., 264, 267, 268, 298 

Kamaran Islands, 290 

Kanagy,J. R., 181, 298 

Karachi, 284 

Karlsbad, 281 

Kasan, 104 

Katathermometer, 27 

Kaunas, 69, 282 

Kelly-Edwards, E. J., 270, 298 

Kendrew, W. G., 108, 298 

Kenya, 71, 151, 153, 290 

Kessler, O. W., 264, 267, 268, 298 

Key West, 72, 123, 287 

Khamsin, no, 140, 161 

Khartoum, 72, 140, 141, 291 

Kimberley, 291 

Kincer, J. B., 208, 210, 215, 218, 227, 

298 
Kingston, Jamaica, 73, 147, 288 
Koepang, 285 



INDEX 



309 



Korea, 71, 135, 285 
Kraus, E. B., 260, 298 

Labrador, 48, 54, 166 
Lacquer, 176, 187 
Lagos, 139, 290 
Lahore, 284 
Lakes 
and house sites, 58 
water supply, 41 
Land breezes, 152, 154 
La Paz, 73, 288 
Lapworth, H., 40, 42, 298 
Larose, P., 253, 298 
Las Palmas, 292 
Latvia, 55, 69, 282 
Leather, effect of humidity, 181 
Lee, D. H. K., 253, 298 
Leicester, Atmospheric Pollution, 190, 

195 
Leipzig, 69, 281 
Lemons, H., 210, 212, 253, 298 
Leningrad, 70, 105, 283 
Lenkoran, 283 
Leopold, L. B., 198, 295 
Leopoldville, 289 
Lerwick, 280 
Leste, no 
Levante, 1 1 o 
Leveche, no 
Lhasa, 285 
Libreville, 289 
Libya, 290 
Lighting, 76, 117, 250 

effect of buildings, 78 
snow, 78, 212 

fluorescent, 251 

natural, 250 
Lightning, 213 

flashes to earth, 215 

rods, 262 
Lima, 289 

Limestone, corrosion of, 189 
Line-squall, 221 
Lisbon, 69, 282 

Liu Kiu Islands, 137, 138, 284 
Lithuania, 69, 282 
Load, 

electric, 251, 252 

thermal, 242 
Locomotives in desert, 162 
London, 

climate, 280 

daylight, 252 

fog, 42, 43 

frost, 81 

gustiness, 86 

humidity, 248 

Meteorological Office, 36, 39, 80, 
223, 235, 298 

pollution by smoke, 192, 195 



London — continued 

radiation, 62 

rainfall, 40 

sunshine, 68 

temperature, 67, 98, 235 

thunderstorms, 205 
Los Angeles, 128, 198, 287 
Lourenco Marques, 290 
Luanda, 289 
Lugano, 70 
Lull in wind, 85 
Lux, 76 
Lwow, 282 
Lyons, 69, 281 

Mackintoshes, 254 

Madagascar, 141, 290 

Madeira, no, 292 

Madras, 71, 284 

Madrid, 70, 282 

Malakal, 141 

Malaria, 152 

Malaya, 22, 27, 71, 157, 285 

Malta, 70, 282 

Manaos, 288 

Manchester, 68, 280 

Manchuria, 71, 135, 285 

Mandalay, 283 

Manilla, 73, 285 

Manley, Gordon, 44, 298 

Markham, S. F., 245, 249, 298 

Markovo-on-Anadyr, 285 

Marseilles, 109, 281 

Marsh, A., 188, 299 

Marshall, W. A. L., 67, 299 

Martinique, 288 

Massawa, 289 

Mauritius, 292 

Mazatlan, 286 

Medan, 285 

Medicine Hat, 125 

Mediterranean, 20, 108, 112, 22: 

Meetham, A. R., 190, 195, 299 

Melbourne, 73, 112, 114, 291 

Melling, C. T., 251 

Menado, 283 

Meuse Valley, fog, 94 

Mexico, 144, 286 

Miami, 287 

Middleton, C. A., 89, 299 

Midowicz, W., 161, 299 

Milan, 282 

Mildew, 180, 182, 184, 187 

Mississippi floods, 123, 208 

Mistral, 109 

Mogadiscio, 290 

Moisture, absorption of, 181 

content of air, 17, 293 
Mojanga, 142 
Mollendo, 164 
Mombasa, 290 



310 INDEX 

Mongalla, 291 
Mongolia, 285 
Monsoon, 132 

East Indies, 156 

European, 104 

Indian, 130 

West African, 138 
Montevideo, 289 
Montreal, 286 
Montserrat, 147 
Morel, Gh., 264, 299 
Morocco, 290 
Moscow, 70, 104, 283 
Mossamedes, 289 
Mosul, 284 
Mould, growth of, 180, 182, 184, 

187 
Mountain climate, 18, 61, 163 

sickness, 163 
Mozambique, 143, 290 
Mud rain, 229 
Mukden, 71, 285 
Muller-Hillebrand, D., 216, 299 
Munich, 69, 281 

Nagasaki, 71, 137, 138, 284 

Nageli, W., 270, 271, 299 

Nana, 137, 284 

Nairobi, 71, 290 

Nanking, 70, 136, 283 

Nassau, Bahamas, 147, 287 

National Smoke Abatement Society, 

190 
Nautical Almanac, 77 
Navigation, interruption by ice, 48, 

53 
Nemuro, 137, 284 
New Caledonia, 292 
Newfoundland, climate, 118 

ice off, 49, 53 
New Guinea, 292 
New Orleans, 72, 123, 124, 287 
New South Wales, 114, 201, 291 
New York, 72, 214, 287 

Institution of Radio Engineers, 216, 

299 
Metropolitan Insurance Co., 214, 

299 
New Zealand, 30, 73, 114, 292 
Niagara, electric, 262 

Falls, 34 
Niamey, 289 
Nice, 69, 281 
Nicosia, 280 
Nigeria, 290 
Night blindness, 256 
Niigata, 137, 138 
Nile Flood, 141 
Nome, 286 
Normanton, 291 
Nortes, 145 



North America, 115, 286 
Norway, 104, 282 
Noumea, 292 
Nyasaland, 153, 290 
Odessa, 283 
Ogive, 236 
Oilskins, 254 
Okhotsk, 105, 285 

Sea of, ice, 51 
"Old Wives' Summer," 104 
Oran, 289 
Oslo, 282 
Otomari, 137, 284 
Ottawa, 72, 286 
Oulu, 54, 281 

Oxford, Institute of Agricultural En- 
gineering, 36, 299 
Ozone and corrosion, 184 

Padang, 285 

Paint, 179, 189 

Pakhistan, 284 

Palermo, 282 

Palestine, 285 

Palma, 282 

Pampas, 149 

Pamperos, 149, 221 

Panama, 11, 72, 154, 201, 287 

Papagayo, 145 

Papeete, 292 

Paper, effect of humiditv, 12, 181 

Para, 288 

Paraguay, 148, 289 

Parde, M., 210, 299 

Pardy, A. A., 270, 299 

Paris, 69, 103, 281 

Parker, A., 188, 299 

Patagonia, 129 

Permafrost, 167 

Pernambuco, 288 

Persia, m, 284 

Perth, W.A., 73, 114, 291 

Peru, 156, 289 

Peshawar, 284 

Philadelphia, 287 

Philippines, 73, 156, 157, 201, 285 

Photophobia, 256 

Phthisis, 144 

Phu-Lien, 71, 284 

Pitch, Flow of, 180 

Pittsburgh, 197 

Plateaus, climate of, 126, 165 

Plum rains (Japan), 138 

Poland, 69, 282 

Pollution, atmospheric, 32, 188 

Advisory Committee, 190 

and lighting, 190 

distribution, 190 

diurnal variation, 196 

effect on buildings, 79, 189 
sunshine, 190, 191 



INDEX 



3" 



Pollution, atmospheric — continued 

measurement, 190, 195, 196 

minimising, 198 

suspended, 196 
Ponta Delgada, 292 
Poona, 59, 174 
Poorga, 106 
Porches, sleeping, 58 
Port au Prince, 146, 287 
Port Blair, 283 
Porter, L. C, 213, 299 
Port Moresby, 292 
Portland, Oregon, 72, 287 
Porto Rico, 147, 225, 288 
Porto Novo, 139 
Portsmouth, 280 
Portugal, 69, 282 
Powell, E. B., 251 
Powell, R. W., 204 
Power, 

hydro-electric, 33, 99, 108, 165 

load, 251, 252 

solar, 38 

wind, 34 
Prague, 68, 281 
Pressure, 

decrease with height, 163 

wind, 88, 89, 164 
Punta Arenas, 288 

Quebec, 286 

Queensland, 114, 201, 291 
Quetta, 283 
Quito, 73, 164, 288 

Radiation, 

actinic, 256 

as a source of power, 38 

effect of slope, 60 

frost, 264 

heating effect, 60, 172 

in different latitudes, 63 

infra-red (long- wave), 57, 63 

scattered, 63 

solar, 18, 57, 172 

transmission coefficient, 59 

ultra-violet, 57, 176, 190, 256 
Rails, temperature of, 175, 178 
Railways, protection from snow, 47, 

127, 212 
Rain, 

and corrosion, 189 
wind, 94 

effect on outdoor occupations, 83 

heavy, 83, 201 

making, 258 

penetration by, 94 
Raincoats, 254 
Raindays, 280 
Raindrops, terminal velocity, 84 



Rainfall, 

and water supply, 38 

duration, 83 

effective, 83 

increase with height, 82, 164 

intensity, 83 

reliability, 33 
Ramdas, L. A., 176, 299 
Rangoon, 283 
Red Sea, 27, 290 
Reitschels, H., 235, 299 
Relf, E. F., 211, 295 
Reservoirs, 41 
Reykjavik, 73, 282 
Rhodesia, 71, 290 
Richards, B. D., 205, 207, 210, 299 
Riga, 69, 282 
Rio de Janeiro, 154, 288 
Rivers, 

and water supply, 41 

discharge of, 207 
Roberts, B., 167, 299 
Roberts, W. O. J., 218, 220, 296 
Rockets and hail, 261 
Rockhampton, 114 
Rome, 69, 282 
Roof, 

colour, 59 

insulation, 75 

snow load, 84 

wind suction, 88 
Rotting, 184 

Roumania, 55, 70, 201, 282 
Rowen, J. W., 255,299 
Royal Meteorological Society, 251. 

298 
Rubber, 176, 179 
Ruby, Col., 261, 299 
Run-off, 210 
Rusting, 183, 187 

Saghalien, 107, 284 

Sahara, 139, 159, 174 

Saigon, 284 

St. John, N.B., 286 

St. John's, Newfoundland, 53, 286 

St. Lawrence River, ice in, 49 

St. Louis, 72, 218, 287 

St. Lucia, 147 

St. Thomas Island, 139 

Salina Cruz, 145, 286 

Salisbury, Rhodesia, 71, 290 

Salonica, 281 

Salt Lake City, 72, 287 

Salt nuclei, 200 

Samoa, 157, 292 

Sandakan, 283 

San Francisco, 72, 128, 287 

San Jose, 287 

San Juan, 288 

San Luis (Argentine), 149 



312 INDEX 

San Salvador, 287 

Santiago, Chile, 73, 150, 288 

Santis, 107 

Saturation deficit, 17, 23, 160, 293 

Schaffer, R. J., 189, 299 

Schiller, P., 253, 300 

" Schlagregen," 94 

Schonland, B. F. J., 214, 300 

Schulze, W. M. H., 176, 181, 300 

Scirocco, no 

Scotland, see Britain 

Sea breezes, 152, 154 

Seasons, 95 

Seattle, 287 

"Seeding" of clouds, 259 

Sewage plants, 199 

Seychelles, 292 

Shanghai, 70, 136, 283 

Sherlock, R. H., 93, 300 

Siam, 131, 285 

Siberia, 

climate, 105 

debacle, 52 

snow cover, 45 
Sierra Leone, 27, 139, 290 
Siesta, no 
Silk, artificial, 32 
Simaika, Y. M., 298 
Simla, 284 
Simoom, no, 161 
Simpson, Sir George, no, 214, 300 
Singapore, 71, 285 
Sitka, 127, 286 
Skylights, and snow, 78, 212 
Smith, J. A. B., 180, 300 
Smith, T., 250, 300 
"Smog," 197 
Smoke, 

dissipation, 91 

from chimneys, spread of, 88, 188 

see also Pollution, atmospheric 
"Smokes," 152 
Smyrna, 285 
Snow, D., 180, 182, 300 
Snow, 43, 164, 212, 359 

cover, 43, 78, 120 

effect on lighting, 78, 212 

heat conductivity, 82 

load on roofs, 84 

sheds, 47, 127, 212 

storms, 100, 120, 123, 212 

water supply from, 127 
Snow-mobile, 168 
Sofia, 68, 280 
Soil, 

conductivity, 82 

"heaving," 82 

temperature, 59, 176 
Solar radiation, see Radiation, solar 
Solomon Islands, 292 
Somaliland, 290 



Soudan, 72, 291 
South Africa, 72, 142, 391 
South America, 148, 154, 288 
South Australia, 114, 291 
Southerly Bursters, 112, 221 
South-west Africa, 142, 291 
Spain, 70, 282 
Spitsbergen, 166, 292 
Springs, water-supply from, 40 
Squalls, 88, 157, 321 
Squires, P., 260, 298 
Stagg,J. M., 62, 64, 300 
Stalker, E. A., 93, 299 
Stockholm, 70, 282 
Stone, R. G., 44, 247, 300 
Storm-wave, 124, 133, 225 
Stornoway, 280 
Strasbourg, 69, 281 
Stuttgart, 281 
Sucre, 288 
Sugar, 181 

Sulphur di-oxide, 188, 196 
Sultriness, 23, 246 
Sumatra, 285 
Sumatras, 222 
Sun, 

altitude, 59, 61, 64 

as a source of power, 38 

azimuth, 64 

declination, 61 

exhaustion, 256 

helmet, 256 

hour-angle, 64 
Sunburn, 256 
Sunshine, duration, 67, 109, 128 

effect of smoke, 190, 191 
Sunstroke, 246, 256 
Surigao, 285 
Sutton, O. G., 92, 300 
Suva, Fiji, 292 
Sverdlovsk, 104, 285 
Swatow, 137 
Sweat, 22, 30, 161 
Sweden, 54, 104, 282 
Switzerland, 27, 70, 282 
Sydney, N.S.W., 73, 112, 114, 291 
Syria, 109, 285 

Tahiti, 292 

Taihoku, 70, 137, 284 

Tallinn, 281 

Tamatave, 290 

Tampa, Fla., 123 

Tananarive, 290 

Tanganyika, 153, 291 

Tannehill, I. R., 122, 227, 228, 300 

Tartu, 68, 281 

Tashkent, 285 

Tasmania, 30, 112, 114, 292 

Tay Bridge, 100 

Teheran, 284 



INDEX 



313 



Tehuantepecer, 145 
Telephone wires, 47, 101, 213 
Telkes, M., 243, 300 
Temperature, 

accumulated, 235 

and electric load, 251 

and humidity, 184 

average, 23 

coefficient, 180 

comfortable, 233 

decrease with height, 23, 164 

effect of cloud, 57 
colour, 171, 174 

effective, 242, 244 

extreme, 174, 176, 242, 280 

frequency, 236 

high, 18, 160 

in container, 1 72 
sun, 18, 171, 174 

of body, 246 
room, 240 

wet-bulb, 23, 26, 245 
Temporales, 150 
ter Linden, A. J., 58, 240, 300 
Textiles, 32 
Thames 

floods, 100 

flow, 209 

ice in, 99 
Thelm, W., 94, 300 
Thermometer, 

black-bulb, 175 

wet-bulb, 23 
Thiessen, A. H., 121, 296 
Thomas, A. M., 89, 300 
Thorman, G. L., 176, 242, 296 
Thorshavn, 281 

Thunderstorms, 121, 123, 153, 155, 
156, 205 

darkness in, 252 

squalls, 222 
Thursday Island, 114, 291 
Tibet, 285 
Tiflis, 283 
Timbuktu, 289 
Time, equation of, 64 
Timor, 285 
Titanic, loss of, 49 
Tobacco crops, 123, 181 

industry, 32 
Tokyo, 71, 137, 284 
Topees, 256 
Tornadoes, 91, 218 

Africa, 140, 221 

Australia, 112 

Britain, 221 

U.S.A., 123, 218 
Toronto, 72, 286 

Towns, atmospheric pollution, 190 
Trail, B.C., smelting works, 91, 198 
Trade Winds, 37, 147, 156 



Transjordan, 71, 285 
Transport, interference by 

dust, 228 

fog, 42 

frost, 48 

glazed frost, 213 

sea ice, 48 

snow, 43, 212 
Transvaal, hail, 211 
Trees as air filters, 79 

wind breaks, 90, 270 
Trincomalee, 158 
Trinidad, 148, 288 
Tripoli, 160, 290 
Trombe, F., 38, 300 
Tromso, 282 
Tropical climates, 151 
Tsientsin, 135, 283 
Tsinan, 135 
Tuberculosis, 144 
Tulagi, B. S. I., 292 
Tundra, 18, 105 
Tunis, 291 

Turbulence, 85, 92, 197, 199 
Turkey, 71, 283, 285 
Twilight, civil, 77 
Tynemouth, 280 
Typhoons, 137, 138, 157, 219, 226 

Ubacs, 107 

Uganda, 154, 291 

Uleaborg, 54, 281 

Ultra-violet radiation, 57, 176, 190, 

256 
United States, 

climate, 115, 286 

Department of Agriculture, 298 

dust storms, 228 

floods, 121, 123 

fog, 43 

glazed frost, 48 

hail, 210 

heat-stroke, 22 

hurricanes, 133, 226, 228 

ice-storms, 48 

lightning, 214 

rainfall, 116, 201, 203 

snow cover, 44 

sunshine, 72 

tornadoes, 218 

Weather Bureau warnings, 
flood, 208 
forest fires, 215 
hurricane, 228 

wet-bulb, 26 

wind, 119, 124, 127 
Uruguay, 148, 289 
Urumtsi, 160, 285 
U.S.S.R., 

climate, 105, 283, 285 

dust storms, 229 



314 INDEX 

U.S.S.R. — continued 

ice, 50, 52, 54, 55 

sunshine, 70 
Utrecht, 69, 104, 281 

Valdivia, 288 
Valparaiso, 288 
Vapour pressure, 1 7, 293 

and corrosion, 185 
Varnish as protection, 187 
Vegetation, effect on pollution, 79 
Venezuela, 73, 155, 289 
Venice, 69, 282 
Ventilation, 30, 58, 243 
Vera Cruz, 286 
Verandas, 58 
Veranillo, verano, 145 
Verkhoyansk, 105, 285 
Vernon, W. H. J., 180, 184, 188, 300 
Victoria, Australia, 113, 114, 291 
Victoria, B.C., 72, 286 
Victoria Falls, 34 
Vienna, 68, 280 
Viscosity, 180 
Visibility, 43 
Vladivostok, 70, 285 

Wallace, W. K., 212, 300 
Walls, 

damage by frost, 80 

heat conductivity, 59, 74 

radiation on, 62 
Walvis Bay, 291 

Ward, R. de C., 119, 123, 124, 300 
Warsaw, 69, 104, 282 
Washington, D.C., 72, 85, 287 
Washington, Bureau of Standards, 1 74, 

263, 301 
Washington, Weather Bureau, 208, 215, 

228 
Water, 

and prevention of frost, 265, 268 

mains, freezing, 81 

power, 33, 99, 108, 165 

repellance, 255 

super-cooled, 47, 259, 261 

supply, 38, 161, 168 

vapour, condensation, 187 

waste, disposal of, 42 
Weather, 

cycles, 103 

forecasts, 103 Zagreb, 283 

recurrences, 101, 117 Zanzibar, 153, 291 

Weeks, J. R., 234, 301 Zomba, 290 

Wegener, A., 221, 301 Zurich, 70, 282 



Wellington, N.Z., 73, 292 
Wells, 40, 148 
artesian, 1 1 3 
West Africa, 139, 221 
West Australia, 114, 291 
West Indies, 73, 146, 226, 287 
Wet-bulb temperature, 17, 23, 26, 245, 

293 
Wheat, damage by hail, 210 
Whirlwinds, see Tornadoes 
White Sea, ice, 50 
White squall, 22 1 
Willy-willies, 112, 219, 226 
Wilmington, 123 
Wind, 

and clothing, 253 

and rain, 94 

breaks, 270 

chill, 167 

cooling effect on buildings, 75, 243 

downdraft, 92, 271 

effect on body temperature, 30, 253 
corrosion, 188 

maximum speed, 218 

power, 34 

pressure, 88, 89 

structure, 85 

suction, 88 

temperature of, 76 

variation with height, 34 
Windhuk, 291 
Windmills, in, 148 
Windows, 

double, 58 

effect on temperature, 58, 243 

lighting by, 250 
Winnipeg, 72, 286 
Winters, severe, 98 
Woodruff, 256 
Woollen industry, 32 
Wright, H. L., 196, 301 
Wright, N. C, 180, 182, 183, 300, 301 

Yap, 292 
Yarmouth, 280 
Yarnell, D. L., 203, 301 
Yokohama, 138 
Yugo-Slavia, 70, 283 
Yukon, 126 
Yuma, 287 



Date Due 
Due Returned Due Returned 


OCT 1 fi '70 


SEP ? a 7$ 






lov i ff V* 










OCT 2/ '7| 






NOV 2 2 76 


RH *v w rt ,- .*, 






FEB i-XI 


F2S 1 4 yf/ 


tfR **31 




MAY t s 71 


' 






AUG 30 '80 


f|flJG 2. i '8U 






W 1 


10V 2 3 '80 






MAB . • 


MAR 2 3 %l 






f : " *f 


■ 






FEB 2 1 1981 


fEB 1 2 1982 






FfeB 2 3> » 


S 






mar 1 .? ...rat., 








"%*Fn» 


SEP 1 8 884 




















































































/ 








\ / 








/ 







liMjaS'BLOf FLORIDA 



3 12L,2 043737A7 2 



337. 5 

ARCHN 

TECTURg 



firs, 



,r »M 



• 



^3 



%